FEDERAL COURT OF AUSTRALIA

Davies v Lazer Safe Pty Ltd [2018] FCA 702

ORDERS

THE COURT ORDERS THAT:

1.    The claim and cross-claim each be dismissed.

2.    Subject to any further order the reasons for judgment are not to be published except to the parties and their legal representatives before 12.00 noon, 25 May 2018.

3.    Any party seeking an extension or variation of order 2 is to file and serve any interlocutory application seeking such relief together with any affidavit evidence to be relied upon by 12 noon, 25 May 2018.

4.    The applicants file any materials in relation to costs or other orders within 14 days.

5.    The respondent file any materials in relation to costs or other orders within a further 14 days.

6.    Unless the Court otherwise orders, costs and/or any other orders, if any, be determined on the papers.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

Table of contents

MCKERRACHER J:

1.    INTRODUCTION

1    The applicants, Mr Kevin Stephen Davies and Triclops Technologies Pty Ltd (ACN 117 350 464), claim infringement of Australian Standard Patent No. 2003229135 for the invention entitled ‘A Safety System’. The Patent is Annexure A to these reasons.

2    The Patent is for an invention relating to a safety system, in particular, for use with machinery having moving parts, such as press brakes. A press brake is a mechanical tool used for bending material, usually metal. It operates with substantial force. The safety system is to detect the presence of an obstruction in the path of a moving part in such machinery. As the Patent discloses, various mechanisms have been used in the past to prevent the operators of industrial machinery, such as press brakes, from placing their hands beneath the moving tool during operation. Such an action by the operator has the potential to cause injury, given the force applied to the tool and the speed at which it can move.

3    An example of a previous safety mechanism noted in the Patent is a physical guard placed between the operator and the tools. The use of physical guards can, however, obstruct the view of the operator during use and/or impede access to the work making it difficult for the operator to perform the job and the task in the most efficient manner. Another example provided is a light curtain which surrounds the path of the moving tool. Light beams are projected onto detectors, which are able to sense when the beam is broken and trigger either a halt or slowing down of the movement of the tool. Again, the disadvantage of the individual light beams is said in the Patent to be that operators are kept away from the work area. Another method of providing safety protection is the use of one or more light beams along the leading edge of the tool, which are arranged to move with the tool and slow or stop the tool if an obstruction breaks the beam. In such arrangements, the beam must be deactivated as the tool approaches the work and the light beam must be set at sufficient a distance from the tool to stop the tool in time. The minimum permissible distance of the beam from the tool is therefore dependent on the maximum speed and movement of the tool. Again, it is suggested in the Patent that a problem with this type of arrangement includes the need to realign light beams when tools are changed and the possibility that an operator may move a hand under the tool just as the beams are deactivated.

4    The Patent is for an invention which attempts to overcome, in part, some of the disadvantages identified in previous safety systems for detecting the presence of obstructions in hazardous areas around machines having moving parts.

5    Mr Davies is the owner of the Patent. Triclops is and has been since 8 October 2015, the exclusive licensee of the Patent.

6    The respondent (Lazer Safe Pty Ltd (ACN 090 113 572) has, since at least 20 June 2015, offered for sale and kept for sale the systems (the Lazer Safe Systems) known as:

(a)    an LZS-005;

(b)    an IRIS; and

(c)    an IRIS Plus (or IRIS +).

7    The main difference between the three laser systems is that the IRIS and IRIS Plus systems contain additional functionalities to the LZS-005. IRIS has additional functionality known as ‘bend speed management’ (BSM). IRIS Plus also has BSM, together with an additional functionality discussed in more detail below. A Product Description was filed by the respondent detailing the Lazer Safe Systems and their functionalities (discussed in, and annexed to Appleyard #2 (defined at [29]) and contained in the Confidential Annexure to these reasons, but which is omitted from publication).

1.1    The dispute and outcome

8    The applicants say that the respondent has infringed the Patent by the sale of its products. The respondent rejects the infringement allegation and asserts in response, for various reasons, that the Patent is invalid.

9    The applicants press infringement of claims 1-5, 8, 12, 14, 21, 22, 28-31, 38, 44 and 46. However, if claim 1 is found to be infringed then, given admissions made by the respondent, the dispute as to infringement of the dependent claims only concerns whether claims 2, 3, 4, 5, 14, 21, 38 and 46 are also infringed (the Disputed Claims). The respondent cross-claims for invalidity.

10    For reasons which follow, in my view, the applicants’ construction of the claims of the Patent is unrealistically broad. On the proper construction, I consider there is no infringement by the respondent’s systems. However, equally, on the proper construction, the Patent was not invalid as asserted by the respondent.

1.2    Overview

1.2.1    The Patent

11    The ‘Summary of the Invention’ (p 3 lines 6-19) constitutes a consistory clause, referable to claim 1. This is a broad description of the nature of the invention, and does not limit the invention to one particular form (the words used are ‘[i]n accordance with one aspect of the present invention…’: p 3 line 6). The consistory clause identifies that the invention comprises three components being:

(a)    ‘at least one light emitting means’;

(b)    ‘at least one light receiving means’; and

(c)    ‘a processing and control means’.

12    Any obstruction within the region illuminated by the ‘light emitting means’ will result in, for example, a shadow being cast on the light receiving means. This affects the information output by the ‘light receiving means’, which then acts upon by the ‘processing and control means’, which may, as a consequence, control movement of a part of the press brake, such as the tool. The invention is described (p 3 line 20 to p 12 line 16) ‘by way of example’ with reference to the drawings accompanying the Patent. The ‘light emitting means’ (as described in the example) serves the purpose of illuminating a region around the path of a tool in order to detect obstructions in the region. That component is used in conjunction with at least one ‘light receiving means’. The ‘light receiving means’ is said to comprise an aperture in a screen and an image detection device (p 6 lines 7-8). Examples given of the ‘image detection device’ are a ‘charge coupled device’ (a CCD) and a projection screen (with a camera to observe the projected image) (p 6 lines 8-9). Both of these types of image detection devices contain an array of light sensors, wherein each sensor outputs an electrical signal based on the level of light it receives. Different embodiments of the light emitting means and the light receiving means are described at p 5 line 1 to p 6 line 22.

13    The ‘processing and control means’ receives ‘information’ from the light receiving means. An issue will be whether ‘information’, as the applicants say, is a broad notion, not limited to, for example, ‘images’. Various embodiments are described at p 7 line 1 to p 11 line 18. The role played by the ‘processing and control means’ commences with an example describing the comparison of ‘images’ stored in the memory of the processing and control means with images received from the ‘image detection device’ (p 7 lines 4-14). That is, ‘image information’ constitutes ‘images’ in this embodiment. This embodiment may be described for present purposes as an ‘image comparison’ embodiment. Another embodiment concerns the creation of ‘a total picture made up of the image information’ received by the light receiving means (p 8 lines 1-7). This ‘total picture’ is referred to as a ‘shadow map’ (p 8 lines 4-5).

1.2.2    Common ground and issues

14    It is common ground that each of the Lazer Safe Systems possesses each of the following integers of claim 1 of the Patent:

15    It is also accepted that if any of the Lazer Safe Systems possess all of the integers of claim 1 of the Patent, that system also possesses all of the integers of claims 8, 12, 22, 28, 29, 30, 31 and 44 of the Patent.

16    Therefore, the following are essentially the issues in dispute in the proceedings:

(1)    What is the proper construction of the Disputed Claims, particularly as to integer 1.5 and integer 1.6 of claim 1?

(2)    Whether each of the Lazer Safe Systems possesses the following integers of claim 1:

(1)    It is also in dispute whether each of the Lazer Safe Systems possess the following integers of dependent claims:

(2)    It is also in dispute whether IRIS and IRIS Plus possess the following integers of dependant claims:

17    In relation to the respondent’s allegation of invalidity, it is common ground that the filing date of the Patent was 10 June 2003.

18    The Patent also asserts a claim to priority from each of the following three provisional applications:

    PS 2878    11 June 2002 (as to which, see [25(5)] below

    220950078    10 July 2002

    2002951351    12 September 2002

19    There are several issues in dispute in relation to invalidity.

20    The first is the date which each of the Disputed Claims is entitled to claim as its priority date pursuant to s 43 of the Patents Act 1990 (Cth).

21    In connection with novelty, it is agreed that:

(1)    US Patent No. 4772801, in the name of Cybelec SA (Cybelec Patent), was made publicly available on 20 September 1988.

(2)    European Patent Application EP 1 246 148 A2 in the name of SICK AG (SICK Patent), was made publicly available on 2 October 2002.

(3)    The SICK Patent can only be relevant for the purpose of novelty if the Disputed Claims are not entitled to claim priority from any of the provisional applications (that is, the priority date of the Disputed Claims is not earlier than the filing date).

(4)    Claims 3, 4 and 14 do not lack novelty in light of the Cybelec Patent.

(5)    Claims 3, 4, 14 and 21 do not lack novelty in light of the SICK Patent.

22    In relation to novelty, the disputed issues are:

(1)    Whether any of claims 1, 2, 5, 8, 12, 21, 22, 28, 29, 30, 31, 38, 44 or 46 of the Patent lack novelty in light of the Cybelec Patent.

(2)    If the SICK Patent is relevant to the question of novelty, whether any of claims 1, 2, 5, 8, 12, 22, 28, 29, 30, 31, 38, 44 or 46 lack novelty in light of the SICK Patent.

23    There is also a claim in relation to inventive step or obviousness in connection with invalidity. It is agreed that:

(1)    The Cybelec Patent was made publicly available in United States of America on 20 September 1988.

(2)    The SICK Patent was made publicly available on 2 October 2002 in the German language.

(3)    The SICK Patent can only be relevant for the purpose of inventive step if the priority date of the Disputed Claims is not earlier than the filing date.

(4)    For the purpose of assessing inventive step, the relevant version of s 7 of the Patents Act predated the Intellectual Property Laws Amendment (Raising the Bar) Act 2012 (Cth).

(5)    The common general knowledge in the relevant art is the same, whether considered at any of the filing date or the filing date of any of the provisional applications.

24    The respondent has provided the applicants with a schedule of matters asserted to form part of the common general knowledge in Australia as at both 11 June 2002 and 10 June 2003 (the respondent’s CGK Schedule). Where the applicants agree that any of those matters form part of the general knowledge as at these dates, the agreement is indicated in the applicants’ response (applicants’ response to CGK Schedule).

25    The issues in dispute in relation to inventive step or obviousness in connection with invalidity are:

(1)    The identity, qualifications, experience and attributes of the hypothetical skilled worker in the relevant art as at the priority date of the Patent.

(2)    Whether any of the other matters asserted to be common general knowledge in Australia in the respondent’s CGK Schedule, but not agreed in the applicants’ response to CGK Schedule did form part of the common general knowledge in Australia as the priority date of the Patent.

(3)    Also in dispute is whether the invention as claimed in any of the Disputed Claims would have been obvious to the hypothetical skilled worker in the relevant art in light of the common general knowledge as it existed in Australia as at the priority date of the Patent.

(4)    Whether:

(a)    the Cybelec Patent could reasonably be expected to have been ascertained, understood and regarded as relevant by the hypothetical skilled worker in the relevant art before the priority date of the Patent; and if so

(b)    the invention, as claimed in any of the Disputed Claims, would have been obvious to the hypothetical skilled worker in the relevant art in the light of common general knowledge as it existed in Australia as at the priority date of the Patent considered together with the Cybelec Patent.

(5)    If the Disputed Claims are not entitled to claim a priority date earlier than the filing date, whether:

(a)    the SICK Patent could reasonably be expected to have been ascertained, understood and regarded as relevant by the hypothetical skilled worker in the relevant art before the priority date of the Patent; and if so

(b)    the invention as claimed in any of the Disputed Claims would have been obvious to the hypothetical skilled worker in the relevant art in the light of common general knowledge as it existed in Australia as at the priority date of the Patent considered together with the SICK Patent.

26    There is also a ‘fair basis’ argument. The issues in dispute are whether claim 1 and all claims dependent upon claim 1 are fairly based on the disclosure of the Patent insofar as they claim the following, being integer 1.5 and integer 1.6:

a processing and control means arranged to receive image information from the light receiving means and thereby recognise the presence of one or more shadowed regions on the light receiving means cast by obstructions in the region;

wherein the illumination of the region is such that the processing and control means has sufficient image information to determine the boundaries of the or each shadowed region and control movement of the part dependent on said image information.

27    There is also a claim based on lack of clarity in relation to invalidity. The issues in dispute in this regard are:

(1)    Whether the use of the word ‘recognise’ in claim 1 is unclear; and

(2)    Whether the use of the phrases ‘image information’ and ‘sufficient image information’ in claim 1 is unclear.

28    I address the issue of infringement, and the issues arising in relation to invalidity, in turn.

2.    THE EVIDENCE

2.1    The key experts – an overview

29    The infringement evidence relied on was:

(a)    for the applicants by affidavits of Mr John Acheson sworn on 27 April 2017 (Acheson # 2) and sworn on 5 June 2017 (Acheson # 3); and

(b)    for the respondent by affidavits of Mr Robert Appleyard sworn on 17 February 2017 (Appleyard # 2) and Mr Peter Berry sworn on 16 February 2017.

30    As to invalidity, the evidence relied on was:

(a)    the respondent’s evidence in chief by:

(i)    affidavit of Mr Appleyard sworn on 17 October 2016 (Appleyard # 1);

(ii)    affidavit of Mr Mark Pullen sworn on 17 October 2016;

(b)    the applicants’ evidence in answer: affidavit of Mr Acheson sworn on 17 February 2017 (Acheson # 1);

(c)    the respondent’s evidence in reply: affidavit of Mr Appleyard sworn on 28 April 2017 (Appleyard # 3); and

(d)    the applicants’ evidence in answer: affidavit of Mr Acheson sworn on 14 June 2017 (Acheson # 4).

31    In relation to claim construction and the issue of infringement, I preferred the respondent’s witnesses, principally, Mr Appleyard, despite the superior qualifications and perhaps more general expertise of Mr Acheson. While Mr Acheson was more qualified generally, he had very limited experience in the particular field of this technology. I am conscious in arriving at this preference that Mr Appleyard was not independent. Despite this, he had very considerable internationally relevant experience and skills and I found his infringement evidence plausible and realistic, despite his obvious interest in the matter. On that invalidity evidence of a more general nature, as distinct from infringement, I preferred the evidence of Mr Acheson. I considered that his evidence on invalidity in response to the respondent’s claims was realistic and objective and to be preferred.

Mr Appleyard

32    Mr Appleyard is a director and principal of the respondent. He has had extensive ‘hands-on’ experience with press brake safety systems since about 1995 when he started designing and building safety systems for press brakes. In 1998, he established the respondent’s business developing safety products for use with press brakes. He was the principal designer of the respondent’s safety system known as the Lazer Safe LZS-003 System, a precursor of the LZS-005. The respondent operates worldwide, in the design and manufacture of press brake safety and control devices. Mr Appleyard has also, since 2001, been involved in writing safety standards for press brakes for the European Union and, since 2009, for the United States. He and the respondent have won various Western Australian awards for engineering excellence, industry and export since 2003. The respondent has received State and Federal funding grants since 1998 for the development of its business.

Mr Acheson

33    Mr Acheson was initially giving evidence for the applicants only as to validity, principally in answer as to the novelty and obviousness of the Patent. He is an American systems engineer based in Iowa. Since 2004, he has worked in the United States principally as a systems designer in aerospace, communications, software and defence. Between about 2002 and 2004, he worked as an intern and then part-time for Accurpress America in Rapid City, South Dakota, whilst completing his studies at the South Dakota School of Mines and Technology. His work with press brake safety systems was undertaken as a project for his studies. Mr Acheson’s exposure to press brake safety systems is quite limited: in time, in nature, in scope and geographically. His involvement with these systems centred on his part-time student design project for his engineering studies, which began in about mid-2003 and after the latest priority date. His experience is American, rather than Australian or European. His evidence discloses no substantial personal awareness or knowledge of this field at an international or Australian level at any of the priority dates. It is notable that the relevant field or art is quite restricted and confined, especially in Australia. Safety systems for press brakes are a small but sophisticated market, centred on Europe (with considerably smaller markets in the United States and even more so in Australia at the relevant time in 2002/2003). Mr Acheson expressly accepted that the United States was a ‘backwater’ for press brake safety systems in 2002/2003.

Mr Berry

34    Mr Berry is an independent electronics and software engineer. He gave evidence on infringement only, in answer to Mr Lim’s proposed evidence on behalf of the applicants. Mr Lim was then not called. Mr Berry holds a Bachelor of Science in Electronic Engineering from the University of Natal (now the University of KwaZulu-Natal) in South Africa in 1983. In 2000, he moved to Perth where he has continued to work as a self-employed engineering consultant and contractor. He has designed and implemented electronic software components and systems for over 30 years, particularly in the ‘embedded’ area where electronics and software are used to control devices and machines. He has performed work for the respondent on an ad hoc basis since about 2006, but such work is only a small proportion of his overall business. He was not involved in work relating to the specific safety features of the Lazer Safe Systems. He was not as qualified as Mr Acheson, but his evidence was helpful and plausible.

Mr Pullen

35    Mr Pullen is a patent attorney acting for the respondent in this dispute. His evidence was of a formal nature exhibiting the various documentary prior art and other relevant documents, together with a history of the application for the Patent and the publication of various prior art documents. He was not required for cross-examination.

2.2    Evidence in chief - detail

36    Before considering the arguments advanced by the parties on the construction of the Patent, I propose referring to the detail of the evidence in chief of the key witnesses, mainly on construction, but partly on infringement and invalidity. I have not relied on the infringement evidence or even comparisons as an aid to construction. The summary does not discuss the cross-examination which, when relevant, is examined in discussion as to the parties’ contentions on each of the aforementioned issues.

Mr Acheson

37    Mr Acheson is the principal systems engineer at Rockwell Collins, a United States multinational company. Rockwell Collins is engaged in the design, production and support of systems for use predominantly in the aerospace and defence industries. Mr Acheson holds a Bachelor of Science and Electrical Engineering and has had 10 years’ experience in control system design. In 2004, he completed his Bachelor of Science and Electrical Engineering in South Dakota School of Mines and Technology in Rapid City. Between the second and third years of study, in around mid-2002, he undertook a summer internship with Accurpress. Accurpress was a manufacturer of press brakes and shears for forming sheet and plate metal. After that internship, he was employed by Accurpress on a part-time basis during his next year of study. At this time, he worked on a few small research and development projects. One of the projects involved development of an angle sensor for measuring the extent to which a piece of metal had been bent by a press brake. This would facilitate partial bending of a piece of metal, that is, not using the full stroke of a machine to bend a sheet of metal to the maximum bend angle, usually 90 degrees. He undertook a variety of administrative and network administration tasks.

38    In the final years of his study, around mid-2003, as part of a final year design class, Mr Acheson undertook a design project to design and build a press brake safety system. The project was undertaken at the request of Accurpress. Accurpress funded the project and provided him with materials and support for it. This was, in effect, an extension of the work he was already performing for Accurpress.

39    After graduating in 2004, Mr Acheson commenced fulltime employment with Rockwell Collins, where he worked for some seven years. During this time he was a member of the Government Systems, Advanced Sensors and Navigation Department. His work focused on digital signal processing for communication and navigations systems and the development of signal processing algorithms. He was heavily involved in work on software defined radios. It is a type of radio where many of the parts and components, typically implemented by hardware, are instead implemented by software. It allows the radio to be reconfigured, utilising different types of signals.

40    In 2011, Mr Acheson joined a company known as Cambridge Silicone Radio. At this company, he was part of the software development of low powered GPS chips, intended for use in applications such as smart phones. After two months with Cambridge Silicone Radio, the director of software for whom he was working, left the company to join a start-up company called ‘mCube’. Mr Acheson also joined mCube at that time, working on algorithms for a navigation system for use indoors. This could be used, for example, in shopping centres or other large indoor public spaces without the need for GPS. He was named as an inventor on two patents which were filed while he was with mCube. At around 2012, the office of mCube that he was working with was spun-off into a different start-up company. He left to join Raven Industries, where he worked as an architect on precision agriculture systems. He was the lead architect on the development of numerous systems. One was a grain yield monitoring system for combine harvesters, providing accurate real-time yield maps. He also worked on the integration of sensors into GPS and automated steering systems for agricultural machinery. Whilst with Raven Industries, he was named as an inventor on eight filed patents. In about April 2016, he left Raven Industries and returned to Rockwell Collins, where he remains employed. He has a number of duties there. He is a systems analyst, working on airborne and ground based software defined radios. He is also the modelling and simulation group leader engineer and leads several software development teams.

41    Specifically, in relation to press brake safety systems, Mr Acheson was, whilst working with Accurpress, engaged to design and build a prototype press brake safety system. This came about as a result of being approached by the owner of the company, Mr Hilton. There had been some incidents with the company’s press brakes and Mr Hilton wanted Mr Acheson to look at designing a safety system to deal with this. Accurpress, at that stage, did not sell press brakes with manufacturer-fitted systems. Some of Accurpress’ dealers offered to install third party light curtain systems for customers. In simple terms, he explained a light curtain creates a vertical ‘curtain’ of light in front of a press brake which can detect objects that enter the light and stop the press brake. Light curtain systems typically operated in the following way:

(a)    two vertical columns were mounted a short distance, approximately one foot in front a press brake, with one column at either end of the press brake;

(b)    one column had a series of vertically spaced apart light transmitters to transmit beams of light across the front of the press brake, while the other column had a corresponding set of receivers, receive the light from the light transmitters; and

(c)    if one or more of the beams of light were broken during operation, the press brake could be stopped.

42    Otherwise if Accurpress customers wanted safety systems fitted to their press brakes, they would have to obtain aftermarket safety or guarding systems and have them refitted to the systems. At this stage, one of Accurpress’ major competitors Trumpf, offered customers the option of having a light curtain system installed in their press brakes. At the time he began working on the project, Mr Acheson had been working with Accurpress for close to two years for an average of about 15 hours per week. As such, he said he had a good knowledge of the operation of press brakes.

43    Mr Acheson described a press brake as being a large piece of industrial machinery, designed to bend sheet metal. Typically, it does this by lowering a long tool or blade onto the sheet of metal to be bent and driving the sheet into a V-shaped die. The press brake pushes the metal all the way into the die to create a complete bend or only part of the way to create a partial bend. Mr Acheson was also aware of the likely safety issues that could arise for an operator of a press brake. The vast majority of press brakes were manually fed. This meant that an operator would stand next to the machine, insert the metal to be bent in the press brake and hold an end of the piece out of the path of the tool to keep the piece in place while bending occurs. The major safety risk to the operator was that he or she could inadvertently insert a part of his or her body, such as a finger, hand or arm into the path of a tool. The tool is relatively sharp and descends with great force, so if a part of an operator’s body became trapped beneath a tool, it could cause serious injury or even death.

44    Another safety risk can arise when the tool pushes the sheet metal into the die to bend the sheet. This causes the edges of the sheet to fold upwards and towards the machine in a V-shape and it is possible for the edge of the sheet to strike an operator in the head or the face as this occurs, particularly if the bending occurs at high speed. It is also possible for a part of an operator’s body to become trapped between the sheet and the side of the blade as this bending takes place. This could be the operator’s hands or fingers if the operator is holding the sheet as it is bent.

45    The first step Mr Acheson took in the project was to look at the existing market to see what types of safety systems were available. He was already aware of light curtains.

46    Mr Acheson undertook internet searches to find out what systems were available. At this time, Google was not as widespread in internet searching. He used a program known as MetaCrawler, being a service that aggregated results from a number of different early internet search engines that were available at the time.

47    He had no knowledge at that stage of whether or not any patents existed and did not undertake any searches to ascertain this. The only systems he uncovered in the searches he did conduct were the various forms of light curtains.

48    Mr Acheson expected the task to be a simple one, namely, taking an existing light curtain, developing the necessary program and applying it to an Accurpress press brake. However, it became apparent that there was a more complex problem. That was how to avoid the system being falsely triggered. By ‘falsely triggered’, he referred to a situation where the light curtain was interrupted, but without actual risk to operator safety. That could occur, for example, if the piece of metal being bent obstructed one of the light beams. This was more common where either a thick piece of metal was being bent or where a sheet was being bent into a box or similar shape in a number of operations. When a sheet was being bent into a box, the previously bent up sides of the box would present a wide profile and would be more likely to block a light beam from the light curtains and trigger a stop. At the time of conducting his inquiries, Mr Acheson was aware that this problem had been approached by varying the threshold required for a sensor of the light curtain to register as being blocked.

49    Mr Acheson understood at the time that the sensors in light curtains comprised a photodiode. A photodiode is an electrical component converting light into an electrical current. The sensors could work in one of two ways:

(a)    the sensor could be set to output either a high/low or on/off signal, depending on the amount of light detected. The amount of light required to switch between signals could be adjusted by a physical adjustment of the sensor; or

(b)    the sensor could output a signal of a particular voltage, depending on the amount of light detected. This was sent to a controller which could be programmed to generate an ‘on’ or ‘off’ signal, depending on whether the voltage was above or below a defined level.

50    In either case, the result was same. A binary on/off or high/low signal would be generated, depending on whether the level of light received by the photodiode was above or below a set level.

51    If a light beam was obstructed by a very thin object, for example, the edge of a sheet of metal only several millimetres thick, some light would still reach the sensor. However, if the light beam was obstructed by a larger object, the amount of light reaching the sensor would be much lower. By varying the threshold voltage at which the signal was triggered, which in turn depended on the amount of light received, the systems could be adjusted so that they did not trigger if only a very thin obstruction blocked the path of the light beam.

52    Another problem with light curtains was that the light transmitters and receivers had to be very precisely aligned. This was to ensure that the light beam, when not obstructed, was detected by the transmitter. If there was a slight misalignment, this could result in either some or all of the narrow light beam missing the receiver. This, in turn, could cause the output of the light sensor to fall below the threshold and trigger a signal, causing the press brake to stop.

53    Mr Acheson largely worked on this project alone. After reviewing the existing market for safety systems, which seemed to comprise various arrangements of light curtains, and seeing the drawbacks of light curtains, he decided to take a different approach. He considered that one way of dealing with the alignment issues would be to use a camera, which would monitor the path of the descending blades. This offered, he said, a fairly robust solution without the types of alignment issues referred to above. As long as the field of the camera covered the relevant area, it would not matter if there was a slight misalignment of the camera. There was therefore greater tolerance if the camera was not perfectly aligned.

54    The camera used at the time was a complimentary metal oxides semiconductor (CMOS) camera and is one type of technology used in cameras. The other type is a CCD. Both CCDs and CMOS cameras perform the same basic function, that is, capturing light and converting it to an electrical current.

55    Mr Acheson thought about using the camera to try and distinguish between the body of an operator and other, permitted objects, such as the workpiece, however, he quickly dismissed this as being too difficult and complex a problem requiring too much processing power.

56    Instead, he applied an optical filter to the camera and worked on distinguishing between a permitted obstruction, such as the workpiece, and an impermissible obstruction, such as part of the operator’s body, based on the intensity of light detected. In order to make a clear distinction between the intensity of light detected from a permitted obstruction and an impermissible obstruction, he used a glove coated in phosphorescent chemical, which was to be worn by an operator, and exposed this to a UV light source. When the glove was exposed to the UV light source, it emitted a high intensity of light. When this level of intensity was detected by the camera, it would trigger a signal to stop the press brake. However, the presence of other obstructions in the field of the camera, for example, the workpiece, die or tool, would not trigger such a signal. In this way, the system would only be stopped if the gloved hand of an operator entered the field of the camera. He developed a working prototype of this system, but the system was never put into commercial use as far as he knew. He believed that the requirement for the user to wear a particular type of glove made this system too cumbersome for practical use.

57    Mr Acheson said he was well aware in mid-2003 of light curtains systems where the light emitting and receiving means are fixed in front of the press brake. As a result of the work he performed on press brakes, he maintained an interest in this field after completion of the project. He became aware of systems which illuminated an area around the blade at a later time. This was one or two years after completing the project. He was not aware of those in mid-2003 and would not have regarded them as being common general knowledge, either at 11 June 2002 or mid-2003.

58    Mr Acheson said he was aware of processing and control means as a common feature of light curtain systems. He would have regarded this as common general knowledge, both at 11 June 2002 and mid-2003. However, the manner in which each system processes information, that is, the exact logic it uses, usually comprised confidential proprietary information and so would not have been immediately apparent. He would therefore not have regarded that information as common general knowledge.

59    On the detection of shadows cast by obstructions between the light emitting and receiving means, Mr Acheson said that if this is referring to the function of a light curtain detecting whether a shadow has been cast upon a light sensor, leading to a corresponding drop in output voltage from the sensor, he was aware of this and considered it was common general knowledge at both dates.

60    As to technology which could be used to record and store the images of shadows cast in the region between the light emitting and receiving means, including shadows cast by the blade, workpiece or an anvil, or an obstruction in the region, Mr Acheson said he was aware of image recording technology generally, such as CMOS cameras. However, at the time he undertook the project he had not heard of its use to record and store images in a press brake safety system. He did not regard this as being common general knowledge at either date. He ultimately used a CMOS camera in the press brake safety system he designed in around 2003. He did not, however, use the CMOS camera to record or store images. Rather, he arranged it to observe an area below the blade of the press brake and trigger if the camera detected an intensity of light corresponding to the coated glove of an operator’s hand.

61    With regard to programming the processing and control means such that it could be used with such technology to determine the boundaries of the shadows cast in the region, Mr Acheson was not aware of the use of such technology in press brake safety systems. It was not, in his view, common general knowledge at either date.

62    As to programming the processing and control means such that it stored images and could compare them with other images, similarly, Mr Acheson was not aware of the use of this technology and did not regard it as being common general knowledge at either date.

63    As to controlling, including stopping the descent of the blade, depending on detecting shadows cast by obstructions, detecting the shape of shadows cast by obstructions, or detecting the shape of shadows cast by obstructions and comparing them with the shape of other images recorded by the system, Mr Acheson was aware that light curtains operated by controlling the descent of blade based on detecting a shadow cast by an obstruction. But none of the systems he was aware of detected the shape of the shadow or controlled the blade based on such information (emphasis added). That was not, in his view, common general knowledge at either date.

64    As to the following systems in the respondent’s CGK Schedule, the Cybelec System, the TIROPS System (standing for Travelling Infra Red Operator Protection System), the Fiessler AKAS System and the Nuova System, Mr Acheson had not come across systems by those names and did not regard them as common general knowledge at either date.

The respondent’s Lazer Safe LZS-003 System

65    Mr Acheson said he recalls hearing of the respondent’s systems at the time he undertook the project. He did not recall a particular system name or model number he was aware of at the time and could not recall specifics of the systems, but his recollection was that they were light curtain systems which were generally well known. As he could not recall specifics of the Lazer Safe Systems at the time, he does not regard those systems as likely to have been common general knowledge at either date.

The Patent

66    Mr Acheson could not recall ever seeing the Patent before he was provided with a copy of it by the applicants’ solicitors. He was given various tasks in relation to the Patent.

67    He expressed the following views in relation to it.

Prior systems identified in the Patent

68    The section in the Patent headed ‘Field of the Invention’ sets out a number of prior systems and the problems that arise with each of the systems in a self-explanatory way.

69    Mr Acheson said that as at the priority date, which he was instructed by the applicants’ solicitors to regard as 11 June 2002:

(a)    physical guards and tethers had been used. They had the problems described in the Patent. Mr Acheson does not regard them as being commonly used as at the priority date because of the problems. He regarded the light curtain systems as being well known at the priority date;

(b)    Mr Acheson was aware of light curtains and the ‘false triggering’ problem referred to in the Patent. He was not aware of any issue with small body parts being able to be passed through a light curtain. While this would theoretically have been possible, in his experience, the problem did not arise in practice for at least one of the following reasons:

(i)    the light curtain was positioned far enough away from the machine so that a finger could not reach the hazardous area of the machine. Reaching a hazardous area would require the operator to pass at least a part of his or her arm through the light curtain. The beams were closely enough spaced that an operator’s arm could not pass between light beams; or

(ii)    alternatively, if the light curtain was close enough to the machine that a finger could reach a hazardous area to pass through the curtain, the light beams would be positioned closer together to prevent this from happening.

(c)    as at the priority date, Mr Acheson was not aware of light beams projected along the leading edge of the tool, nor the problems associated with them.

General comments on the Patent

70    Mr Acheson said he found the Patent to be clearly written and had no trouble understanding the Patent or the invention it disclosed. It is a safety system for a press brake. It has a light transmitter and a receiver which receives light from the light emitter. The light emitter illuminates an area that is included in the path of the descending blade.

71    He said the light beam that is described in the Patent has a two dimensional cross section (that is, it is not a single point or a flat beam). The Patent described a lens arrangement at the transmitting end of the beam that is used to spread a point beam and then collimate the beam by making the rays of light that comprise the beam substantially parallel to one another so that the beam will have zero or negligible spread (that is, the cross section of the beam will not vary or vary only by a very small amount over its length).

72    At the receiver end, the Patent described a way of collapsing the light beam back to a smaller size. The light receiver also includes an image detection device. A processing and control means processes the images captured to search for unknown shadows and can stop or slow movement of the tool depending on what is detected. The Patent also mentioned that the processing and control means may use ‘shadow maps’ which are prior recorded images of what a safe view looks like as the tool moves through its path of movement. The system can compare these shadow maps with the shadows being cast on the receivers as the tool descends.

73    Mr Acheson’s view was that the invention described in the Patent is not like any systems he had seen or heard of as at the priority date or in 2003. Had he been aware of this Patent at the time he was engaged in the design of the safety system, it would have been of great interest to him and, in particular, its ability to distinguish between safe and unsafe obstructions.

Claims of the Patent

74    Mr Acheson referred to the ‘claims’ commencing at p 13 of the Patent and comprising 47 numbered paragraphs, all being for a ‘safety system’. Only claim 1 is a stand alone claim and the remaining claims add features to claim 1 or other previous claims, which ultimately trace back to claim 1. Having read the Patent, most of the words or phrases appearing in the claims of the Patent were clear and he could understand them based on his skill and present experience. He considered that he would have similarly been able to do so as at the priority date. The applicants’ solicitors asked him to provide an understanding of the meaning of certain words and phrases appearing in the claims. In relation to them, he said the following.

‘Image information’ in claim 1

75    Mr Acheson said he understood ‘image information’ to refer to the information that constitutes the ‘image’ that is detected by the image receiver; this is the image that is provided to the processing and control means. He understood that an image to be constituted by the output from one or more sensors, together with the details of its location. He considered that an image can exist where there is only one output from one location. As such, the information will be ‘image information’ as long as it includes the output of at least one sensor together with its location. It could also be the output from an array of sensors with information about the position of each sensor within the array.

‘Recognise the presence of one or more shadowed regions’ in claim 1

76    Mr Acheson took recognition to be referring to the ability to detect shadows cast by multiple independent objects. It requires the ability to ‘recognise the presence’ of a shadow that is cast. The reference to ‘presence’ he said implies a concept of time, that is, that there is an ability to know at any given time whether a shadow is there or not.

‘Obstructions in the region’ in claim 1

77    In this context, Mr Acheson understood an obstruction to be any opaque object which blocks the light travelling between the transmitter and the receiver. That could be, he said, fingers, hands, the workpiece, the anvil or the tool itself.

‘Processing and control means has sufficient image information to determine the boundaries of the or each shadowed region’ in claim 1

78    Mr Acheson said this referred to the required resolution of the image information and he understood the phrase to be saying that the resolution of the image information that is received by the processing and control means must be sufficient to allow determination of the boundaries. He noted that the phrase does not mention that the processing and control means must carry out boundary determinations. Rather, it only requires that the processing and control means have the necessary information to do so.

79    He also said that in order to understand what resolution would be required for boundary detection, it is helpful to understand how ‘boundary detection’ or ‘edge detection’ is carried out. He said it is one of the cornerstones of image analysis. He was familiar with edge detection, having applied it in projects on which he worked while at Raven Industries. While this was well after the priority date, he regarded the basic concepts as having been unchanged since around the 1970s.

80    ‘Edge detection’, he said, involves image information being analysed to search for a transition from a ‘dark’ output in one pixel to a ‘light’ output in an adjacent pixel or vice versa. In the case of photodiodes, this would involve a comparison of the voltage output of adjacent photodiodes looking for a sufficient step or difference, referred to as a gradient between the outputs. ‘Edge detection’ is a term Mr Acheson used interchangeably with ‘boundary detection’. I will refer to ‘boundary detection’ to keep the language consistent with the Patent.

81    Therefore, to have sufficient resolution to carry out boundary detection, there must be enough pixels that the transition from ‘light’ to ‘dark’ can be seen. As this phrase in claim 1 refers to ‘boundaries (plural) of the or each shadowed region’ (which are cast by one or more obstructions in the region) (emphasis added) there must be an ability to determine multiple boundaries on potentially multiple shadows. Mr Acheson regarded that requirement as being met if boundaries can be identified sufficiently to give an impression of the shape and location of a shadow, albeit in a coarse or grainy manner. As resolution increases, it becomes possible to detect boundaries with more precision. However, a relatively coarse resolution may still allow determination of the boundaries in the sense he described.

‘A single parallel beam of light’ in claim 5

82    Mr Acheson said this phrase requires that there is a single continuous beam of light, rather than using multiple point light sources in an array to illuminate an area. The phrase also requires the beam of light to be parallel, but does not say to what the beam is parallel. He therefore understood the term ‘parallel’ to refer to being parallel to itself, that is, that the rays of light in the beam are parallel to one another, or collimated, and the beams spread or beam angle is zero or very close to zero.

‘Columnated’ in claim 35

83    Mr Acheson treated this term as being a spelling error for what should be ‘collimated’. He was not familiar with the word ‘columnated’.

Acheson # 2

84    In a further affidavit, sworn on 27 April 2017 (Acheson # 2), Mr Acheson reviewed the evidence of Mr Appleyard in Appleyard # 2. Mr Acheson said that, in relation to the Lazer Safe Systems, the systems first scan and then identify the location of the tool and then blank segments representing the location of the tool. This means that the shadow cast by the tool would not stop the tool. Only if a shadow is detected in an unblanked area, that is ‘an unknown’ shadow, is the tool stopped. Mr Acheson spoke of an additional possible function for the Patent, which he understood as being as follows:

(a)    a processing and control means may create ‘a total picture’ made up of the image information received as a tool moves through its path of movement, that is, a ‘shadow map’;

(b)    the shadow map is a series of images, in effect, a video of what is observed as the tool passes through its full path of movement;

(c)     in the first instance, it will include the tool and the anvil only, this is created on a first pass of the tool, with no workpiece present; and

(d)    subsequently, additional shadow maps may be created which include images of the workpiece being bent.

85    It is only the final stage of the second function, namely the creation of shadow maps, including the workpiece, that is described by Mr Appleyard. Mr Acheson said this particular aspect of the operation of the system, being the comparison with the pre-stored images of the workpiece being bent, was specifically identified in the Patent.

86    In relation to integers 1.5 to 1.6, Mr Acheson disagreed with the view expressed by Mr Appleyard that none of the Lazer Safe Systems has integer 1.5 or integer 1.6. Mr Acheson said Mr Appleyard ‘seems to assume’ that these integers impose a range of requirements based on examples in the Patent of how the invention could operate. Mr Acheson did not see those as being requirements of claim 1 of the Patent. Mr Appleyard’s views on the matter, Mr Acheson said, seem to be influenced by:

(a)    his view that the tool is not an obstruction as referred to in integer 1.5 and integer 1.6. A view with which Mr Acheson disagreed; and

(b)    that the Lazer Safe Systems are not ‘safety systems’ after guarding mode is fully muted and the purpose of the BSM is to enhance productivity. Mr Acheson disagreed with this, saying the two functions described, namely, BSM and ‘Bendshield Plus’, which operate after the guarding mode is muted, provide a safety function. Systems having these functions, in Mr Acheson’s opinion, are ‘safety’ systems when these functions are active.

87    In relation to integer 1.5, Mr Acheson’s disagreement with Mr Appleyard was based on two assumptions that appeared to have been made by Mr Appleyard:

(a)    that ‘image information’ meant information on the shape of an obstruction; and

(b)    that ‘recognising the presence of a shadowed region’ required comparison with a stored image.

88    Mr Acheson disagreed that integer 1.5 imposed either of those requirements.

89    In relation to integer 1.6, again he considered that Mr Appleyard’s views were based on the assumption that integer 1.6 required that the system both:

(a)    determined the shape of any obstruction; and

(b)    controlled movement of the part based on the determination of the shapes of the obstructions or identification of the obstruction.

90    Mr Acheson disagreed that integer 1.6 imposed either of those requirements. Rather, in relation to integer 1.6, Mr Acheson said the following:

(a)    the processing and control means must have sufficient image information to meet two requirements:

(i)    it must have sufficient image information to determine the boundaries of each shadowed region. That term appears in integer 1.5, and refers to a shadow cast on the light receiving means by an obstruction in the illuminated region. The ‘shadowed region’ may be a shadow of only part of an object. What is likely to happen in practice is that if the hand of an operator starts to enter into the illuminated region, the tip of a finger may enter the illuminated region first. It follows that the first shadowed region from this obstruction to appear in the light receiving means will be a small shadow cast on the edge of the light receiving means by the tip of the finger. It is only the boundaries of the shadowed region that must be able to be determined. It is not the boundaries of the physical obstruction that must be capable of being determined. For example, in a case where the physical obstruction is a hand entering the illuminated region, at the moment it first appears and starts to cast a shadow on the light receiving means, the shadowed region may be only a very small section, such as a single pixel; and

(1i)    it must have sufficient image information to control the movement of the parts, that is, the tool based on that information. This means that the processing and control means is able to generate output signals to control movement of the part based on the image information it is receiving. Nothing in the words of integer 1.6, or the claim when read as a whole, require that the control of the part be based on a determination of the shape or identification of what is the physical obstruction;

(b)    there is no requirement that boundary determination is actually carried out, only that the image information is sufficient to do so; and

(c)    the opening words of integer 1.6 require that the manner in which the region is illuminated allows the quality of image information that is required to be generated. This requires that the light transmitter transmits light with a high enough energy to create sufficient contrast between light and shadow on the receiving means. This allows a gradient to be found between the light and dark pixels as described above. It also requires that the illumination covers the range of sensors, that is, pixels sufficient for boundaries to be detected.

91    Mr Acheson also took issue with Mr Appleyard’s views about ‘obstructions in the region’. The prior description contains statements to the effect the tool was not an obstruction. Mr Acheson’s view was that the tool, and indeed any object that casts a shadow on the light receiving means, is an ‘obstruction in the region’. In the case of the tool, it is an obstruction that is deemed safe and, therefore, the system is able to continue operation, even with this obstruction present.

92    In relation to operation of the Lazer Safe Systems during ‘guarding mode’, Mr Acheson said:

(a)    during set up, diagonal scanning software scans the image of the guarding area at the full camera resolution to identify the lowest point of the tool. Mr Acheson believed it must do so by detecting a boundary, that is detecting a transition from light to shadow at this point; and

(b)    a predetermined number of ‘segments’ around the tool and around the corners of the guarding area are identified so that these can be blanked or disregarded. Certain segments are given a ‘weighting’ which conveyed to Mr Acheson these segments are treated differently in some way. However, this is not explained in the Product Description. Other segments are ‘blanked’, which conveyed to Mr Acheson that these segments are disregarded completely. He said that if there is weighting, rather than blanking, he would require further information as to what the term ‘weighting’ meant.

93    In relation to ‘bending mode’, Mr Acheson noted that the LZS-005 does not operate after the guarding function is muted and the tool begins to bend the workpiece. However, both IRIS and IRIS Plus provide functions during bending of the workpiece. IRIS and IRIS Plus provide the BSM function. IRIS Plus also provides angle measurement of the workpiece being bent. The feature called Bendshield Plus is also described, but is had not been put into commercial use, according to Mr Appleyard.

94    Mr Acheson understood that BSM operates by scanning the image to find a point on the top of a workpiece. This is found by the transition from a light pixel to a dark pixel, detecting a boundary of the shadow cast by a workpiece. As the workpiece is bent, the image is continuously scanned to find the new location of the top of the workpiece. By tracking the position of the top of the workpiece as it rises, the angular velocity of the workpiece can be calculated. Although this occurs after the ‘guarding’ function is turned off. In Mr Acheson’s opinion the system was still functioning as a safety system when BSM is operational. The purpose of BSM is to control the speed of the bending to avoid ‘sheet overhang injury’ as described in the Product Description. IRIS and IRIS Plus differ from the system described in the Cybelec Patent. The Cybelec Patent only provides a safety function, in his opinion, prior to the bending of the workpiece.

95    The bend angle measurement feature of IRIS Plus operates by identifying a straight line representing the edge of the workpiece on either side of the tool. The system then calculates the angle between these lines. The purpose of the function is to check, in real time, whether a target angle has been has been achieved. Mr Acheson did not regard this as a safety function.

96    However, Bendshield Plus involves a guarding function being turned back on for a period of time once bending is commenced. A significant number of segments in the obstruction matrix are blanked and it is only an area between the workpiece being bent and the tool that remains unblanked. This is to protect against an operator getting a finger trapped between the workpiece as it bends upwards and the side of the tool. It is not clear from the Product Description how the system determines which areas are to be blanked when Bendshield Plus is enabled, but Mr Acheson understood that the operation of this function would be the same as described above for guarding mode, thereby, providing a safety function.

97    In relation to location of obstructed segments, Mr Acheson expressed the view that the Lazer Safe Systems must know the location of these segments, contrary the view that expressed by Mr Appleyard, as the tool will only be stopped if an unblanked segment is deemed obstructed.

98    As to information used in BSM, in order to find the relevant pixel each time the system must be constantly scanning to locate the transition between light and dark, the boundary representing the leading edge of the workpiece. This requires analysing the output of all pixels along the scanning path in order to find this transition point.

99    Mr Acheson disagreed with Mr Appleyard’s view that the Lazer Safe Systems did not have the presence of integer 1.5. He said that the processing and control means can:

(a)    during ‘set up’, identify the presence of the tool tip and blank out the segments occupied by the tool;

(b)    during ‘guarding’, determine each segment, either obstructed or unobstructed, based on comparison of the voltages returned in each segment to a threshold; and

(c)    during BSM, identify a point in the top service of the workpiece based on where the output of the sensors transition from light to dark.

100    Mr Acheson considered that it is only if the tool is regarded as not being an obstruction, that Mr Appleyard’s view could be correct. Mr Acheson said, however, that:

(a)    the information available to the processing and control means is plainly ‘image information’;

(b)    each of those processes involve ‘recognising the presence of a shadowed region’;

(c)    the tool is an obstruction and the function of the system in identifying the tool during set up is relevant when considering integer 1.5; and

(d)    the system is a safety system when BSM is operational and therefore the operation of BSM is relevant when considering integer 1.5.

101    Mr Appleyard’s view relied on the assertion that Lazer Safe Systems do not know the location of the obstructed segment. Mr Acheson disagreed, explaining that:

(a)    the process of detection of the tool during set up clearly demonstrates that the processing and control means has sufficient image information to find the tool tip;

(b)    even when considering the only segment information that is passed to the Press Controller and Safety System (PCSS), this would allow the determination of the boundary of any obstructed segment or group of segments simply by checking the status of the adjacent segments;

(c)    the operation of the BSM function of each of IRIS and IRIS Plus plainly requires sufficient image information to determine boundaries, as the process involves determining a boundary;

(d)    the tool is an obstruction and the function of the system in identifying the tool during set up is relevant when considering integer 1.6;

(e)    the system is a safety system when BSM is operational, therefore, the operation of BSM is relevant when considering integer 1.6; and

(f)    in both guarding mode and BSM, the image information is sufficient to be used, and in fact used, to control the movement of the part. In guarding mode, both the image information used to locate and blank the tool, as well as the image information received when the tool has descended, are used to stop the tool if another obstruction, outside the tool obstructed area or any other blanked areas, is detected. In BSM, the image information is used to calculate the bend speed and slow the tool if the speed is too high.

102    Integer 1.6 is present in the Lazer Safe Systems, according to Mr Acheson.

103    Mr Acheson also disagreed with Mr Appleyard in relation to the latter’s view on claims 2, 3, 4, 5, 14, 21 and 46, providing the following reasons (and some commentary on claim 38):

Claim 2

•    I do not agree that the additional integer of claim 2 is not present in the Lazer Safe Systems.

•    The processing and control means stops the movement of the part if an obstruction is detected in an unblanked segment of the obstruction matrix. These unblanked segments are determined during set-up, and are a “predetermined or calculated area” of the region.

Claim 3

•    I do not agree that the additional integer of claim 3 is not present in the Lazer Safe Systems.

•    The processing and control means is able to determine whether the obstruction is in a front, middle or rear area (relative to the tool) (see Product Description page 7). The system can respond differently depending on which area the obstruction arises. For example, when in Box Mode as described in paragraph 35 of the Product Description.

Claim 4

•    I agree that the additional integer of claim 4 is not present in LZS-005.

•    I do not agree that the additional integer of claim 4 is not present in IRIS and IRIS Plus.

•    The BSM function calculates the angular velocity of the workpiece as it is being bent. It then slows the descent of the blade if the angular velocity is too high. This velocity could either be the velocity of just the operator side of the workpiece (relative to a stationary plane), or the velocity of the operator side of the workpiece relative to the other side of the workpiece (which, assuming a symmetrical V-shaped die, would be twice the velocity of the operator side only). In the first case, as the tool is stationary, the calculated velocity would be the velocity relative to the tool. In the second case, the calculated velocity would be of one obstruction relative to another.

Claim 5

•    I do not agree that the additional integer of claim 5 is not present in the Lazer Safe Systems.

•    The opinion expressed here seems based solely on the assertion that the term “relatively large” is unclear. By itself, without any reference to compare the size of the region, this term would be unclear to me. However the claim states that the region is “relatively large with respect to the size of a leading edge of the part” [emphasis in original]. This phrase however is not unclear to me. When I look at, for example, Figure 6 of the Product Description, it is clear that the region includes multiple segments to either side and below the tip of the tool, and that the tip of the tool occupies only a small area of the region. Consequently, I regard it as “relatively large” compared to the leading edge of the part.

Claim 14

•    I do not agree that the additional integer of claim 14 is not present in the Lazer Safe Systems.

•    Paragraph 27 of the Product Description describes that the equivalent of five fully obstructed pixels is required for a segment to be deemed obstructed. The ways in which this can occur are illustrated in RMA-30. In order to have the equivalent of five fully obstructed pixels, an obstruction must have a minimum thickness sufficient to span two pixels (as shown in the first example in RMA-30). This equates to a thickness of approximately 1.33 [millimetres], which is determined based on the pixel size. An obstruction of this size is too small to be part of the operator’s body. If an obstruction is smaller than this, it will not sufficiently obstruct enough pixels in a segment for that segment to be deemed obstructed. As a consequence, the movement of the part will not stop.

Claim 21

•    I do not agree that the additional integer of claim 21 is not present in the Lazer Safe Systems.

•    I am not able to understand the reason for Mr Appleyard’s opinion on this integer. He clearly states a screen is used. Paragraph 76 of the Product Description states that the screen receives the laser light from the sender unit (i.e. the image is projected onto the screen), and the camera observes its pixelated view of the screen (i.e. detects the image information).

Claim 38

•    Based on the information in the Product Description and the Second Appleyard Affidavit, I am not able to either agree or disagree that the additional integer of claim 38 is not present in the Lazer Safe Systems.

•    A lens at the receiver in such a system is used to refract light and focus the beam of light onto the array of sensors in the receiver. This is required where the area of the array of sensors is smaller than the area of the projected light. For example, in a typical domestic camera, a Jens arrangement is used to focus incoming light onto sensors within the body of the camera, which are much smaller than the field of view of the camera.

•    Based on my knowledge and experience of the operation of systems such as this (i.e. having an array of light sensors), the only situation in which I am aware a lens would not [emphasis in original] be used is if the array of sensors is of the exact same physical size as the cross sectional area being observed. In my experience, this would be unusual, since sensor arrays are typically smaller than the area being observed, therefore require a lens to focus the light onto the sensors. I would therefore consider it unusual if the Lazer Safe System did not use such a lens. However without further information regarding the optics of the receiver, I am not able to clearly say one way or the other.

Claim 46

•    I agree that the additional integer of claim 46 is not present in LZS-005.

•    I do not agree that the additional integer of claim 46 is not present in IRIS and IRIS Plus.

•    Plainly the tool is arranged to bend material. Plainly IRIS and IRIS Plus control the tool during bending, for example during the operation of BSM, or BendShield Plus when these features are present.

•    Mr Appleyard’s opinion on this integer seems to be based on the notion that IRIS and IRIS Plus are not “safety systems” when they control the tool during bending. As I describe above, BSM is a safety function and as such, these systems are still functioning as a safety system when BSM is operational.

(paragraph numbers omitted)

104    In relation to Mr Berry’s affidavit, Mr Acheson said that Mr Berry did not provide any clear explanation of what he understood to be required by integer 1.5 and integer 1.6. However, similarly to Mr Appleyard, Mr Berry’s opinion on the presence of those integers, or otherwise, is based on the premise that claim 1 of the Patent requires the type of object identification that is described merely as a possibility in the Patent. He did not agree with that. Mr Acheson said:

While I do not agree with the premise on which Mr Berry proceeds in providing his opinion, I make the following comments on the opinion he expresses:

a)    Mr Berry asserts that the Lazer Safe Systems do not attempt to determine boundaries or relative positions of silhouetted objects in the region, and do not use any form of image comparison. However, the system scans to identify the tip of the tool (i.e. finding a boundary and location). It [t]hen applies blanking based on this location. As described above, I regard this as a form of image comparison, i.e. between the map of blanked segments, and the actual output at a given time.

b)    Mr Berry also asserts that there is no storage of old images. However the map of blanked segments (once created) must be stored in order to be subsequently used.

c)    Mr Berry states that once an obstruction is determined in an unblanked segment, no further information is gathered, and no further processing is performed. This is plainly incorrect. The operation of L-filtering, Hard Shell Soft Centre filtering, and the operation of the latch function all require further processing once an unblanked segment is deemed obstructed. I note however that the Previous Product Description which was provided to Mr Berry in making his affidavit did not include any reference to these functions.

d)    Mr Berry states that determination of boundaries cannot be made based on a single binary value. I agree with this statement. However the Lazer Safe System has more than just the binary value of the obstructed segments.

Acheson # 3

105    In Acheson # 3, Mr Acheson considered a Further Product Description, dated 23 May 2017, which he was provided a copy of by the applicants’ solicitors. Mr Acheson indicated that further elaboration of the Product Description appeared to make clear that the tool tip in the Lazer Safe Systems is detected as:

(a)    the system scans along the diagonal path of pixels, commencing on the top left hand corner;

(b)    it proceeds until it detects the shadow of the tool. It detects the shadow by identifying a transition from a light to dark pixel. The Further Product Description does not allow elaboration on how the system distinguishes between light and dark pixels, however, the description of what is occurring is the same as Mr Acheson had earlier alluded to. Image information is being analysed to identify a gradient, that is a sufficient step or difference between the output of the two adjacent pixels so that an edge of a shadow cast by the tool can be indicated;

(c)    once the point of transition from a light to dark pixel is determined, the system commences a new scan. This new scan commences one row of pixels below the previous scan and will again continue until the shadow of the tool is detected; and

(d)    the above process is repeated until the system scans from one side of the array to the other without detecting a shadow. At this point, the dark pixel in the previous scan is deemed to be the tool tip.

106    It follows that on the first scan, the system will determine the transition point between light and dark pixels corresponding to the left hand of the shadow cast by the tool at a point part-way down the tool. On each later scan, it will determine the transition point between light and dark pixels corresponding to the left hand edge of the shadow cast by the tool, one row below the previous scan. It will continue determining these points until the tool tip is found. The points together represent at least a part of the boundary of the shadowed region cast by the tool.

107    The operation of the Lazer Safe Systems in this manner reinforced his view, Mr Acheson said, that integer 1.6 is present in Lazer Safe Systems for the following reasons:

(a)    the illumination of the region must be such that there is the required contrast between light and shadow. If this is not the case, the system will not be able to distinguish between light and dark pixels;

(b)    the image information available to the systems is sufficient to carry out not just a single scan to identify the tool tip, but multiple scans. Each of the scans determines a point on the boundary of the shadowed region cast by the tool. The process, therefore, results in a determination of at least part of this boundary;

(c)    the image information available to the system must be sufficient to allow the boundaries of any shadowed region to be determined by further scanning of the type carried out during the detection of the tool tip; and

(d)    the scanning to identify the tool tip occurs during the set up of the system. By identifying the location of the tool tip, the system is able to identify the correct segments of the obstruction matrix to blank or disregard while the tool is descending. Therefore, this image information ultimately informs the system whether to respond to a segment that is deemed to be obstructed, that is, by stopping or slowing the tool, or to disregard it because the segment has been blanked. Therefore, the control of the movement of the part is based on this image information meeting the requirement previously described.

108    Mr Acheson also expressed the view that the Further Product Description confirmed to him that all values for every segment in the 15 x 15 matrix are sent to the PCSS, more specifically to the guard counter module in the PCSS. Filtering and muting refer to different things. Both processes are carried out by the guard counter module in the PCSS after receiving the values for every segment in the 15 x 15 matrix. The sequence of operation is still the same as he had previously understood when he swore Acheson # 2. Nothing in the Further Product Description changed the views he had expressed in the Acheson # 2.

Mr Appleyard

109    Mr Appleyard is the director of the respondent and has given both factual and expert evidence.

110    Mr Appleyard left school in 1973, going on to complete a two year course at TAFE and qualifying as an architectural draftsman. By the time he finished that course, his interest had turned to mechanical engineering. He was involved in sailing and yachts, competing at State and national events, and was offered a job as a draftsman and a rigger/mast builder.

111    He worked in this industry and in his mid-20s, having secured commercial production agreements with a builder and additional sponsorships, built a yacht to compete in the World Quarter Ton Cup Championships in Italy, in which he was Australian champion at the time.

112    On returning to Australia, Mr Appleyard worked as a consultant with the yachting industry for about a year. In 1982, he was approached to become the ‘state manager of the spar maker and alloy yacht master builders Alspar’, which wanted to establish a manufacturing branch in Western Australia. He worked in the engineering side for some time and subsequently purchased the Western Australian branch and bought into the Sydney business.

113    Mr Appleyard was involved in engineering throughout this time and was very involved in high level research, development and manufacturing for the America’s Cup yachts. He was involved in designing and developing technologies for these yachts for the Australian campaigns of 1983 to 1987. He was involved in the same way with the Whitbread Around the World racing yachts.

114    He carried this experience from the late-1980s and 1990s in light alloy engineering into other fields. For example, he was involved in design and manufacture of:

(a)    lightweight crane systems for military patrol boats built in Henderson, south of Fremantle;

(b)    alloy and stainless steel products used in the building industry;

(c)    a large span 80 metre mobile irrigation system for the farming industry; and

(d)    concrete form work systems for multi storied buildings.

115    Mr Appleyard first became involved in the industry and technology relevant to press brakes in about 1995. By that time, his father and brother had established a business called Safe-T-Corporation (STC) involved in designing and building safety devices for press brakes. He was involved first as a consultant, but from about 1996, as fulltime manager involved in product development, and subsequently part owner of the business. He left that business in November 1997 and established the respondent’s business in March 1998 for the purpose of developing a safety product for use with press brakes. The respondent had two directors, himself and Mr Ian Costley, an accountant with whom he had previously worked. It was funded by a number of seed investors. The first product was the Lazer Safe LZS-003 System. It was developed by Mr Appleyard over the period from 1998 to 2000 and was CE certified in 2000 (‘Conformité Européene’ or European Conformity certified), which meant that sales of the product could be made in Europe, being the main market for products of this sort at the time.

116    In general and simple terms, Mr Appleyard described press brakes and related safety systems up to 2002 to 2003 as follows:

(a)    The press brake is an industrial machine used to bend sheet metal as part of the manufacturing process of metal products such as fridges and other white goods, metal building products, automotive vehicles and aeroplanes.

(b)    The press brake has an electronic controller, which controls its operations. The human operator first uses the controller to input the requirements for the particular bending operation.

(c)    The operator of the press brake stands in front of it and places the metal to be bent (the workpiece) on the ‘lower tool/die’, also called the anvil. The operator then causes the ‘upper tool/punch’, press tip or blade to descend by using the press brake controls. Typically a press brake will have a foot pedal which is operated by the operator such that it leaves his hands free to position the workpiece.

(d)    The method of operation involves three stages:

(i)    the blade, workpiece and anvil before the blade descends with the workpiece sitting on top of the anvil;

(ii)    the blade as it touches the top of the workpiece; and

(iii)    the blade as it continues into the V of the anvil, causing the workpiece to bend between the blade and the anvil.

(e)    The blade of the upper part of the tool descends until it touches the workpiece. On continuing downwards into the V shape of the anvil, the blade causes the workpiece to bend upwards on either side. When the workpiece has been bent into the appropriate shape, often but not always 90 degrees, the blade descent will stop and will return to the start position. There are some press brake safety systems which work the other way around, with the lower part of the tool moving upwards towards the stationary upper part.

117    Based on this, Mr Appleyard stated that there is a risk to the operator in manually placing the workpiece on the anvil in that the operator may put a hand or finger between the blade and the workpiece as it sits on the anvil. If the descent of the blade is not controlled, there is a risk that the operator could suffer injury, such as amputation to a finger. Given the repetitive nature of the work being carried out, such that an operator might bend several hundred identical workpieces in a shift, there is a serious risk that an operator might lose concentration and make a mistake in the timing and placing of the workpiece.

118    Given the safety concerns, a number of safety systems were developed in the decades between the 1960s to the 1990s. By the time Mr Appleyard first became involved in the industry, one of the common safety systems was a ‘light curtain’, which had a series of light beams placed between the operator and the workpiece. If the operator broke the light beams of the light curtain at the relevant time, the system would send a signal to the controller to the stop the descent of the blade. More sophisticated systems had beams of light emitted and received on either side of the press brake, such that the beam of light was positioned below the leading edge of the blade and moved up and down with it. The emitter and receiver were accompanied by a control system, which controlled it and communicated with a separate control systems in the press brake itself.

119    The beam systems used what is known as the ‘mute point’. That is, when the blade reached the point immediately above the anvil, when it is no longer possible for the operator to place his finger between the blade and anvil because of the distance being too small, that is, between about two millimetres and 10 millimetres depending on the system used, the safety system is ‘switched off’, such that it will no longer react when the beam is broken, although the system itself is still operating. Some systems actually turn off the safety system when the mute point is reached, in this way, by the time the blade touches the workpiece and starts to bend it, the safety system is no longer detecting brakes in the beams. If the safety system continued to operate below the mute point, it would detect the workpiece itself and bending of the workpiece could not be completed. There are a number of complications, however, arising out of the use of beams systems, such as how to allow for a workpiece which has an uneven surface or where the workpiece has already had one side bent.

120    Mr Appleyard said that the respondent established itself as one of the leading companies worldwide in the field of design and manufacture of press brake safety and control devices. This has been, he said, ever since its commencement in 1998.

121    As at October 2016, the respondent employs 35 people in its Perth facility, including 12 in engineering and software, 6 in management and sales, 13 in manufacturing and 6 in administration.

122    In June 2002, the respondent had 14 employees, in addition to Mr Costley and Mr Appleyard, with 4 in management and administration, 6 in engineering and 4 in production.

123    Based on his knowledge of the press brake industry and the market for press brake safety systems, Mr Appleyard estimated that the respondent’s share of the world market in 2016 was approximately 65%. In about 2002, he estimated that the respondent’s market share was then about 30%.

124    In addition to his involvement in the design and manufacture of the respondent’s products, Mr Appleyard has become engaged in a number of committees involved in preparing standards for press brake safety systems. Since 2001, he has worked as a member of the ‘CEN/TC WG 143 Committee’, that is, the ‘Central European Normalisation Technical Committee Working Group 43’. This is the European body that writes the safety standards for press brakes and related machinery for the European Union. Mr Appleyard was on the panel of experts that wrote the latest generation standard EN12622. This was done between 2002 and 2009 and is the current standard in use today. The British standard 12622:2001 from 2001 was an early version of the standard that the Committee subsequently worked on. Based on that experience, Mr Appleyard was invited to join the ‘United States of America committee B.11’. This is the comparable standard writing body for the United States. Again, he was one of the panel of experts involved for writing the current generation press brake safety standard of 2012 for the United States. This work started in 2009.

125    Either the respondent, or Mr Appleyard, have received a number of industry awards for the products developed by the respondent over the years, including:

(a)    winner of Products and Manufacturing category in 2006 WA Engineering Excellence Awards;

(b)    finalist in Small to Medium Manufacturer Export Awards in 2006 West Australian Industry and Export Awards;

(c)    finalist in Small to Medium Manufacturer Export Awards in 2003 West Australian Industry and Export Awards; and

(d)    Australian and Western Region finalist in the Entrepreneur of the Year Award 2006.

126    The respondent has also been the recipient of various substantial State and Federal funding grants over several years, including in:

(a)    1998, $125,000;

(b)    2006, approximately $2.5 million;

(c)    2014, some $2.3 million; and

(d)    eight years of qualification for and claiming of an ‘Export Marketing Development Grant’.

127    The press brake safety system developed by STC in about 1996 became known as the TIROPS System. He was involved in the development of the TIROPS System. It was based on the concept of installing light beams close to the blade, instead of the traditionally accepted practice of light curtains. This involved fitting a light emitter and receiver on either side of the press brake, such that the beam was below the blade.

128    When Mr Appleyard joined STC in about 1995, the TIROPS System product was a collection of standard third party parts, semi-assembled together and installed onto a press brake machine by STC. This was a relatively simple electrical mechanical installation, as would be done by electrical contractors. His father and brother were both electrical contractors.

129    The TIROPS System consisted of infrared light senders and receivers, readily available from a company called Banner. These were connected and therefore driven and controlled by a simple programmable logic controller (PLC). A simple set of logic was written into the PLC, such that when the lights senders were interrupted, a stop signal was given to the press brake. This is a simple operation, which would detect an interruption to any one of the two beams at first two, then later three, and the machine stopped in 100 to 300 milliseconds. This system, with its relevant overrun distance (that is the distance the tool continues before being stopped and the safety system being triggered) was only acceptable on the slow, low productivity machines, mainly in use at the time in Western Australia, that had an operating speed of up to approximately 50 millimetres per second. On faster machines, there was a risk that the system would not react quickly enough and the operator would still suffer injury by having the blade trapping his or her finger between the blade and the anvil.

130    Mr Appleyard was brought into STC to use his product development experience to turn this into a viable product. STC had sold or installed about twenty systems before he joined STC. Mr Appleyard was concerned about the quality and reliability of the early TIROPS System. It was intended to be a safety product, competing against certified failsafe light curtains. The TIROPS System used infrared light, invisible to the naked eye and which could be very divergent and reflective, that is, the light beam spread out over distances and reflected off surfaces such as the metal surfaces of the press brake itself. It was also difficult to align the beam emitter with the receiver. It used a simple single channel PLC with software written by an electrical contractor, not software engineers.

131    By 1995, however, Mr Appleyard set about a complete redesign of the whole concept and created a ‘map’ of what the product should be. He obtained a State government grant of about $50,000, which he used to commission an electronic engineering company called Omnitronics to work for him and STC to assist in its design. He selected laser light to be the light source on the basis that it is visible and can be focussed. The TIROPS System, which he designed, used photodiodes in the receiver. These are devices which generate an electric current on receiving light, the voltage of which can be measured and monitored by the PLC. When a light beam is obstructed, it causes a shadow on the diode (or part of it) which results in the lowering of the voltage which is detected by the PLC. The PLC is programmed so that it will trigger the system to stop the descent of the press brake tool on the voltage falling below a certain threshold, for example, the voltage drops by 50%.

132    Mr Davies was an employee of Omnitronics during the time that this development work was being carried out. He became involved in the development work as an engineer participating in hardware design, including circuit boards and the development of the software. As part of the development work, Mr Davies developed an algorithm which could be used to assist and determine whether a brake in the laser beams was caused by a vibration or an obstruction.

133    On completing this development work in conjunction with Omnitronics, Mr Appleyard established sales and installation of the TIROPS Systems, both in Australia and some regions of Asia. About 100 to 120 of the TIROPS Systems were sold before he left STC in November 1997.

134    In relation to the market for press brake safety systems, when Mr Appleyard first became involved in the STC business and development of the TIROPS System in the late-1990s, he realised the market in Australia was very small, with only about 150 press brakes sold per year.

135    As at June 2002 to 2003, with one exception, there were no businesses in Australia, apart from the respondent, designing and manufacturing safety systems for press brakes. STC was no longer operating. The one exception was a company called Red Eye, which was making light curtain safety systems. It was not a major competitor and went out of business in about 2004.

136    Mr Appleyard realised in the late-1990s that to build a business around a product involving press brake safety systems, it would be necessary to develop a product which could be sold overseas. In particular, he identified the European market as a market with the greatest potential for such a product. The reasons for this were:

(a)    there were, and still were at the time Mr Appleyard swore his affidavit, a large number of press brakes manufactured and used in Europe. In the late-1990s, and after June 2002 to 2003, about 4000 to 5000 machines were built and sold a year;

(b)    a number of the main manufacturers of press brakes in the world are based in Europe, such as Trumpf, Amada, LVD Group and Adira;

(c)    most of the main manufacturers of press brake safety systems and related equipment are based in Europe, such as the German company Fiessler, Pilz, Nuova and SICK;

(d)    the press brakes in Europe tended to be more sophisticated with more demanding safety requirements. Light curtains were used widely on press brakes in Europe;

(e)    the regulatory regime and requirements in Europe, particularly within the European Union were, and still were at the time Mr Appleyard swore his affidavit, more stringent than elsewhere in the world, including the United States and Asia, and put the onus on press brake manufacturers to comply. For example, before a product could be sold in the European Union, it had to meet the requirements of CE certification, which could involve rigorous testing before certification. By June 2002 to 2003, press brake safety systems were required to comply with European CE safety directives and specifically CE standard EN12622, and all other standards that this referenced, to obtain CE certification.

137    In developing both the TIROPS System, and later the Lazer Safe LZS-003 System, Mr Appleyard researched press brake safety systems available in Europe, including the systems produced by Fiessler, Pilz and SICK. He obtained brochures at trade fares and exhibitions and got brochures and documents setting out operating instructions directly from customers and the press brake manufacturers. From 1999, Mr Appleyard started to regularly attend trade shows and exhibitions relating to press brakes, including:

(a)    EMO, a major exhibition for metalworking machinery held once every two years in various countries in Europe;

(b)    EuroBLECH, which was an event in Hanover, Germany, held once every two years, and attended by companies from all over the world involved in sheet metalworking technology;

(c)    Lamiera, held once every two years in Bologna, Italy, attended by companies involved in machines and equipment for machining of sheet metal and related products; and

(d)    Blech Expo, an annual international trade fare for sheet metalworking held at various venues.

138    Mr Appleyard said that press brake manufacturers and their suppliers regularly exhibited at and attended each of, the above trade shows and exhibitions from 1999 to 2002. By attending he was able to meet potential customers and see what competitors in the field were doing. The respondent first exhibited at the EMO exhibition in Paris in 1999 and on a regular basis at various exhibitions thereafter.

139    From 1999 and before June 2002 to 2003, Mr Appleyard also met with press brakes manufacturers in Europe, such as Trumpf, LVD, Amada and Adira to discuss their requirements for safety systems for press brakes.

140    By the time of the development of the Lazer Safe LZS-003 System in 1999 and by June 2002 to 2003, Mr Appleyard considered he had quite a detailed knowledge of the market for press brake safety systems and the available technology in Australia, Europe and the United States. He maintained such knowledge over the years to ensure that the respondent’s products continue to be at the leading edge of press brake safety system technology.

141    There were issues to confront when he designed the Lazer Safe LZS-003 System. The TIROPS System used three laser beams, each produced by a diode, which emits a laser beam intended to have a generally circular cross-section. Typically, the diameter of the laser beam, as emitted, will be about 6 millimetres, however, on very close inspection he noticed that due to the way the laser beams were produced, the cross-sectional shape of the laser was more like a rectangle of approximately 2 millimetres x 4 millimetres. At best, it was a blurred or a distorted circle.

142    Further, when emitted over a distance, the laser beam gradually expanded in diameter or in rectangular size. Typically for the TIROPS System, the diameter expanded to about 10-12 millimetres over a distance of about 5 metres. For the Fiessler AKAS System, this was more towards a diameter of 20 millimetres over 5 metre distance.

143    The receiver for each laser beam in the TIROPS System was a simple diode. It had a sensitive area of about 4 millimetres in diameter. When light covered the sensitive area, that is, from the laser beam, the diode produced a voltage.

144    Mr Appleyard noted the need for there to be a threshold of this level of voltage to determine the detection of an obstruction. This is to allow for inherent vibration in the whole safety system/bending machine installation. Vibration, which causes oscillations of the laser beam, is due to movement and vibration of a very large piece of machinery moving up and down at speed. The level of light received by the receiver may be going up or down or even cut off completely.

145    Another issue identified, is that light only travels in straight lines in a vacuum. When it travels through atmosphere, such as air, it moves around or deflects. This is due to the fact that the air is, in fact, a series of small mini currents of air at different temperatures. These pockets of air therefore have different densities. Cold air is more dense than warm air. Light passing from one density of air to another causes a ‘lensing’ effect, which bends the light.

146    Typically, the metal of the machine, including the bending tools, can be very cold. Then work starts and heaters (for example, the sun or air-conditioners) warm the air in the factory faster than the steel tools themselves can warm. Then the air moves around, particularly around these large moving machines and moving people. The effect can be easily seen because the laser light of most machines is visibly red. It can be seen as a red dot of laser light on the outer casing of the receiver. When a machine is completely at rest, the dot of the laser can be seen moving randomly around. This can be by 1 or 2 millimetres in all directions for a machine 3 metres to 5 metres long or even, in extreme situations, up to 10 millimetres. To combat this, the size of the laser dot was designed to be larger than the receiver area. Also, the voltage setting was typically set so the voltage drop used to detect an obstruction was typically about half of the voltage. Allowing for all tolerances and variables meant that when an obstruction, such as a hand or finger, half obstructed the laser beam receiver diode, this would cause a controlled stop of the machine.

147    The Lazer Safe LZS-003 System did not use multiple laser beams received by a corresponding number of receiver diodes. Rather, it used one large laser beam. It was approximately 2 to 4 millimetres in length and 50 millimetres wide, which Mr Appleyard described as a ‘planar’ laser on the basis that the 50 millimetres width was in one plane. The planar laser was received by one receiving structure that was 40 millimetres wide. This one structure had nine receiving photodiodes spaced equally along the 40 millimetre structure. It meant that any part of the 50 millimetre wide planar laser beam could illuminate any of the diodes.

148    Horizontal vibration did not cause a problem with the Lazer Safe LZS-003 System because there was always part of the beam being detected by the receiver diodes. For vibration in the vertical direction, lenses were placed in front of the receiving diode, receiving structure in the vertical plane, such that any light arriving in a different vertical plane was bent or deflected by these lenses to the receiving diode. But even with the extra refinement of the planar laser beam and use of lenses, in principle the Lazer Safe LZS-003 System worked in the same way as the TIROPS System or the Fiessler AKAS System. That is, if one of the photodiodes was half obstructed or shadowed, such that there was an approximately 50% drop in the voltage output detected for that diode, an obstruction was indicated and the controller would trigger a controlled stop of the machine.

149    In early-2002, the respondent developed a dual laser beam system, utilising two laser beams in parallel below the descending blade. The effect of that system is that the upper beam is positioned 4 millimetres below the blade and the lower beam is positioned 10 millimetres below the blade. In this system, when the lower beam reaches the mute point, it slows the speed of the descent of the blade and mutes. When the upper beam reaches the mute point, the safety system switches off. In this way, a controlled descent of the blade in the deceleration zone is created by the two laser beams.

150    Mr Appleyard produced a copy of the patent for the respondent’s dual beam system based on the provisional application first filed on 27 March 2002. The dual beams systems were sold to Trumpf shortly after the provisional application was filed.

151    In relation to the Fiessler AKAS System, in about 1997, when Mr Appleyard was working with STC, he was carrying out research to see what optical safety systems were available in Europe and first identified the Fiessler AKAS System in documentation obtained at that time. Later, in 1999, he saw the Fiessler AKAS System in operation on machines at the 1999 EMO Exhibition in Paris, where he first demonstrated and displayed the Lazer Safe LZS-003 System. He spoke to the Fiessler representatives and has had regular contact with them since. They sit together on the same EU standards committee. The respondent was competing directly with Fiessler and Mr Appleyard tried to keep up to date with its products and developments. He also discussed the respective merits of the Lazer Safe and the Fiessler AKAS systems with customers in Australia and overseas from 1999 to June 2002 to 2003.

152    Mr Appleyard explained that the Fiessler AKAS System, available in 1999 to 2002, operated in a similar way to the TIROPS System. There were three laser beams below the descending blade of the press brake, but in a different pattern to the TIROPS System. The three beams were received by a receiver unit which had three apertures in it to receive light from each of the beams. Behind each aperture at the back of the receiver unit were two diodes for receiving light. Mr Appleyard noted that the diodes operate in the same manner as the diodes on the TIROPS System and the Lazer Safe Systems.

153    The diameter of each of the laser beams produced by the Fiessler AKAS System was about 6 to 8 millimetres on transmission, expanding to about 20 millimetres when used in large wide press brake. By the time the beam is received by the receiver, the diameter of the light beams would be larger than the aperture. The diodes are fixed behind the apertures at the back of the receiver unit, such that they receive the constant amount of light in the absence of obstructions.

154    The Fiessler AKAS System only protected the zone directly under the tool and in front on the operator’s side. The TIROPS System at the time did this, but also the back edge of the tool as well, that is, on the other side from the operator.

155    While working on the patent application for the respondent’s planar beam system, in about May 1999, Mr Appleyard found various patents in the name of Fiessler. These were provided to him by Watermark, the respondent’s patent attorney, pursuant to search reports commissioned by the respondent or as a result of earlier systems cited against the respondent’s patent application. He studied the Fiessler patents for the purpose of the respondent’s patent application and for gathering information on what Fiessler was doing. He produced patent documentation from Fiessler. The documentation is consistent with his recollection of the Fiessler AKAS System marketed and sold by Fiessler in about 2001. According to his recollection, he read the Fiessler patent in an English translation in about 1999, when working with Watermark on the respondent’s patent application for the planar beam system.

156    What was unique with the Fiessler AKAS System as opposed to the other systems was the use of a camera or other imaging system. None of the earlier systems used a camera with an array of pixels or other imaging device for the purpose of determining the shape of the obstruction. However, when designing the Lazer Safe LZS-003 System in 1999, and in subsequent developments prior to June 2002 or June 2003, Mr Appleyard was aware of a camera and other imaging technology which could have been used with the respondent’s Lazer Safe LZS-003 System.

157    In particular, Mr Appleyard recalled discussing the use of a camera or imaging technology with colleagues, including Mr Paul Sertis and Mr Dimitre Stanev, during the development of the respondent’s dual planar laser beam in about 2001 to early-2002. The technology available at the time, and which was discussed, included:

(a)    camera chips advertised online and consisting of a miniature lens and a processor which could be programmed to record and retain images;

(b)    scanners, similar to barcode scanners, which consisted of a camera chip and a processor;

(c)    CCDs, which typically had at the time, an array of light sensitive cells, like pixels in a camera, each of which would produce a small electrical charge on receiving the light. The charge from each cell could be identified and processed, such that an image was created;

(d)    CMOS digital scanners, which had an array of light sensitive sensors or pixels, with an amplifier for each pixel, enhancing the signal. A product specification produced by Photobit, dated May 2000, was an example of documents Mr Appleyard recalls looking at around 2001 to 2002.

158    The Patent’s first provisional application refers to CCDs as being available, including use of an optical mouse and a CCD camera. It refers to a ‘standard CCD’ in discussing its use of this device. Mr Appleyard knew at June 2002 to 2003 that these systems would have given the Lazer Safe Systems the ability to record images, including images of the blade, workpiece and anvil or any obstruction. However, after looking at the available technology and discussing with his colleagues, Mr Appleyard came to the conclusion that there was no advantage to using a camera or other scanning technology over the use of diodes.

159    In Mr Appleyard’s opinion, at June 2002 to 2003 and at the time of swearing Appleyard # 1, there were no distinct disadvantages in using a camera or other means to determine the shape of the obstruction in a press brake safety system. The following reasons formed the basis of that conclusion:

(a)    the system does not need to know the shape of the obstruction, it just needs to know that there is an obstruction;

(b)    a system that detects the shape of the obstruction is unnecessarily complex involving additional hardware and software;

(c)    a system which detects the shape of an obstruction will take more processing time, which may be critical when the system needs to trigger as soon as an obstruction is detected;

(d)    there is more room for error, given the additional steps involved in detecting the shape of the obstruction and determining whether it is an obstruction which should trigger a safety system; and

(e)    the safety standards which Mr Appleyard was actually involved in drafting did not require it, but emphasised the need to trigger a response as soon as an obstruction was detected.

160    So, none of the respondent’s safety systems have ever used imaging for the purposes of detecting obstructions.

161    Mr Appleyard first became aware of a system produced by Nuova, an Italian company, in about the late-1990s. The first system produced by Nuova was fixed to stands on either side of the press brake; on one side the transmitters and on the other side the receivers. It was positioned to send a single beam above the anvil of the press brake. It was static and did not move with the descending blade, unlike the TIROPS System and the Fiessler AKAS System. Nuova’s system was described as the ‘DFS System’. The instruction manual for the DFS System of 1997 shows a laser beam emitter/receiver. Mr Appleyard believes he saw that manual in the late-1990s.

162    At some time, in about late-2001 to early-2002, the respondent’s press brake customers in Italy, such as Gasparini, made him aware that Nuova had developed a new system. He attended the Lameira Exhibition in Bologna, Italy in 2002, where he saw a press brake safety system being promoted by Nuova, which he presumed was the system which the Italian customer has been referring to. Unlike the earlier DFS System described, this system had the transmitters and receivers moving with the press brake blade, fixed on either side of it, in a similar way to the TIROPS System and the Fiessler AKAS System. Mr Appleyard saw an instruction manual for the Nuova ‘DFS Laser Beam’ System dated 20 January 2004, but its content is consistent with his recollection of what he saw in the system in Bologna in May 2002.

163    The revised Nuova System had a laser beam transmitter and receiver configuration which, based upon the instruction manual, can be described as follows:

(a)    the beam emitted by this system is passed through lenses, such that it forms a block laser beam in the general shape of a flattened letter ‘W’;

(b)    the receiver has a number of receiving elements, specifically diodes, which generally match the shape of the beam; and

(c)    the positioning of these components, the beam and receivers, can be seen in the instruction manual.

164    As with the TIROPS System, Fiessler AKAS System and Lazer Safe LZS-003 System, the Nuova DFS Laser Beam System would be triggered on obstructions casting shadows on a receiving diode, such that a voltage produced by the diode would fall below the set threshold, which Mr Appleyard assumed to be about 50%.

165    Mr Appleyard also discussed the Cybelec System. He cannot recall when he first became aware of it. He recalled reading it when he was preparing the respondent’s patent application for the planar beam system in about May 1999. He believed that it was provided to him by Watermark, having been identified in a search carried out by Watermark, or on the Cybelec Patent being cited against the respondent’s patent application. He was not aware of use of the Cybelec System, either in Australia or overseas at the time.

166    Based on his understanding and knowledge of, and experience in, press brake safety and control systems at June 2002 and 2003, there being no relevant difference in his understanding between those dates or when he first read about it in May 1999, the Cybelec Patent described to him a system which was primarily used as a control system for the purpose of measuring the angle of the workpiece as it was bent. The workpiece is bent by the blade moving downwards into the V of the anvil, each side of workpiece moves upwards, creating an angle between each side. If the workpiece is horizontal, which it usually is, the initial angle is 180 degrees. As the workpiece is bent, the Cybelec System can measure the angle as it changes from 180 degrees to, for example, 90 degrees.

167    Based on the Cybelec Patent, Mr Appleyard considered that at June 2002 to 2003 he would have understood the Cybelec System to work as follows:

(a)    the light being emitted was passed through lenses which expanded into a beam with a circular cross section, the diameter being relatively large;

(b)    the light being received in the embodiment consisted of a rotating box with a number of holes through which the light passes, the light being directed to diodes;

(c)    the system had an angle coder and microcoordinator, or controller, used to process the information, including the light received on the diodes and the position of the box at any time;

(d)    when the box stood still, the light passing through the holes and directed onto the diodes would be seen as dots of light corresponding to the size of the holes. The number of dots would depend on the number of holes. Three holes would produce three dots;

(e)    when the box rotated at speed, the light passing through the holes would be received on the diodes and processed by the controller, such that the system sees a number of semi-circular arcs of light, the number of arcs corresponding with the number of holes;

(f)    as the workpiece bent each side would rise up and cause a shadow to be cast on each arc of light;

(g)    the system can draw a line, in effect, between the shadows cast on each side of the system, seeing two lines, one for each side of the workpiece; and

(h)    this information can then be used to calculate the angle between the sides of the workpiece.

168    The Cybelec System, as disclosed, could have up to twenty holes or diodes. In those circumstances, the distance between the arcs is much smaller and the distance between the shadows cast on the arcs is also smaller. In this way, the full shape of the workpiece could be determined by the system and the angle of bend much more clearly shown.

169    According to Mr Appleyard, all of these operations happen during the bending process and after any safety device has been muted to allow bending to occur. He said that in this mode, the Cybelec System is not a safety device. If twenty holes or diodes were used instead, the Cybelec System would have the capacity to see a number of shadows cast in the arcs of light and to determine the shape of the obstruction.

170    Mr Appleyard has reviewed the Patent and, subject to certain words which he contends are unclear, understands what it claims as the invention. Based on his understanding, knowledge and experience in press brake safety and control systems as at June 2002 and 2003 (though claiming there to be no relevant difference in his understanding between those two dates) the Patent involves a safety system for use of a machine, such as a press brake, with features present and generally known in these earlier systems and this field. He asserted the system has at most as its distinguishing feature over those earlier systems, the capability to recognise the shape of an obstruction and control the machine, depending on the shape of the obstruction recognised.

171    In relation to the wording used in the Patent, Mr Appleyard considered the following words and phrases to convey the following:

(a)    a ‘safety system’ is a system which is intended to protect the user of press brakes from injury caused by, for example, getting a finger trapped between the blade of the tool and the workpiece. This can be differentiated from a control or productivity system, which is concerned with other aspects of the press brake, for example, determining the correct angle of bending of the workpiece, as with the Cybelec Patent;

(b)    ‘image information’ means that the shape of the obstruction must be capable of being determined and sufficient to be used by the system;

(c)     the use of the word ‘recognise’ in relation to the presence of shadowed regions means that the system is not just detecting the shadowed region, but the system recognises the shape of the obstructions;

(d)    a ‘shadowed region’ emphasises that the system can determine the shape of the obstruction;

(e)    an ‘obstruction’ is something which is in the way of the blade of the tool and which should not be there during the descent of the tool towards the workpiece. Neither the blade nor the workpiece can be ‘an obstruction’. A clear example is the operator’s finger; and

(f)    use of the words ‘boundaries of the or each shadowed region’ emphasises that the shape of the obstruction must be capable of being determined and sufficient to be used by the system.

172    Mr Appleyard said:

(a)    the use of the word ‘recognise’, as opposed to ‘detect’, suggests that ‘the presence of one or more shadowed regions on the light receiving beams cast by obstructions in the region’ is recognised, i.e., already known and matched with what is known. However, by its very nature, an obstruction is something which should not be there and not known to the system. For example, the operator’s finger, which is an obstruction and which the system does not have a prior image of. In fact, it is the very fact the obstruction is not recognised, which causes the safety system to trigger and stop the descent of the blade; and

(b)    the phrase ‘sufficient image information’ used in the Patent, according to Mr Appleyard, is inherently vague. It refers back to ‘image information from the light receiving means’, but it is unclear as to precisely what information is being utilised and how much information is required for it to be ‘sufficient’.

173    In relation to the provisional applications, Mr Appleyard said that while the invention described in the Patent in claim 1 is broadly described in the provisional applications, claim 1 of the Patent includes the integer that ‘… the processing and control means has sufficient [image] information to determine the boundaries of the and each shadowed region’. This phrase is not included anywhere in the provisional applications. He could not identify any other text in the provisional applications which broadly describes such an integer.

174    Mr Appleyard asserted that each of the provisional applications describe a system which does more than just detect obstructions. Instead, the system is used to build a library of images, or shadow maps, which images are recognised by the system. The images may be of the tool itself or parts of the workpiece or anvil on which the workpiece is being bent. It is only when something is recognised as not matching these images, that the safety system is triggered. The system detects an image which it compares with the images already stored in the system. If the shape of an obstruction does not match an image stored in the system, the safety system will be triggered.

175    In Mr Appleyard’s opinion, claim 1 of the Patent is wider or broader than that which is described in the provisional applications. Subject to qualifications about the word ‘recognise’, he asserted that claim 1 of the Patent does not require the system to compare the shape of an obstruction with the library of shadow maps, which are what is described in the provisional applications.

176    However, the use of the word ‘recognise’ in claim 1 suggested to Mr Appleyard that the system is comparing the shape of the obstruction with something which is known, that is, the existing library of shadow maps in the system. The system must be doing more than ‘detecting’ the shape of the obstruction, it must recognise it is not one of the shapes already known to it in the library of shadow maps.

177    Therefore, Mr Appleyard said, either:

(a)    claim 1 (and all the other claims which are dependent on it) is wider than that which is described in the provisional applications, in that claim 1 does not require the system to compare the shadow of the obstruction with the shadow maps; or

(b)    the word ‘recognise’ necessarily implies that the system is making a comparison between the obstruction and the shadow maps and recognising that the shadow cast by the obstruction is not the same as one of the shadow maps. Otherwise the provisional applications and the Patent disclosed no way of the system operating by ‘recognition’.

178    The Patent Cooperation Treaty (PCT) application for the Patent, filed on 10 June 2003, Mr Appleyard noted does not use the word ‘recognise’, but instead describes ‘the safety system being arranged to detect the presence of an obstruction’ (emphasis added). Claim 1 of that application does not refer to shadow maps, although it does later refer to ‘shadow regions’.

179    Further, the amendments emphasise the requirement in claim 1 that there be ‘image information’ and added the wording that ‘the processing and control means has sufficient image information to determine the boundaries of the or each shadowed region’.

180    In summary, therefore, Mr Appleyard said:

(a)    the provisional applications are concerned with the use of shadow maps and a system which involves comparing known images with images which the system does not recognise;

(b)    the PCT application uses the word ‘detect’, rather than ‘recognise’;

(c)    the Patent, as granted, uses the word ‘recognise’, but it is unclear whether this means that claim 1 requires a comparison with known images, that is, the shadow maps.

181    For the same reason Mr Appleyard said, as discussed in relation to claim 1 in the provisional applications, claim 1 of the Patent is wider in scope than the specification, which is concerned with describing the use of shadow maps, whereas claim 1 does not refer to shadow maps, unless the word ‘recognise’ and the phrase ‘image information’ are interpreted to refer to shadow maps.

Wording in claim 1

182    Mr Appleyard disagreed with Mr Acheson’s comments on the meaning of ‘image information’ and ‘image’ as used in claim 1 of the Patent. The word ‘image’ meant to Mr Appleyard, in the context of the Patent, the shape or outline of the shadowed region cast by an ‘obstruction’. The phrase ‘image information’ means information relating to that shape or outline. The output of a single sensor may be part of the image information, but by itself, does not provide sufficient information as to the shape or outline to constitute ‘image information’.

183    He disagreed that the workpiece was an obstruction, rather it is the very thing which the blade is intended to bend as part of the normal operation of the press brake. However, if part of the workpiece is already bent, such that it is between the blade and the part of the workpiece to be bent, then part of the workpiece could be considered an obstruction because it appears between the blade and the part of the workpiece to be bent. The anvil is not an obstruction. By the time the blade reaches the anvil, it is already bending the workpiece. The anvil is not between the blade and the workpiece.

184    The tool, or specifically the blade, is not an obstruction, according to Mr Appleyard, by its very nature. The blade cannot be something between itself and something else. The introductory words in the patent specification say ‘the present invention relates to a safety system … to detect the presence of an obstruction in the path of the moving part’. The illuminated region described in claim 1 is in the path of movement of the ‘moving part’. The part itself is not an obstruction.

185    Mr Appleyard disagreed with the suggestion that the processing and control means does not actually carry out boundary determination. To function as described in claim 1, the processing and control means must determine the boundaries of the shadowed region caused by an obstruction. The ‘sufficient image information’ is the boundaries of the, or each, shadowed region, which information is then used to control the movement of the blade.

186    In Mr Appleyard’s view:

(a)    a single point cannot determine an edge or boundary;

(b)    to determine a boundary, there must be multiple points identified, such that the shape or outline of the boundary can be determined;

(c)    identifying points along a single edge of the boundary will not determine the boundary or boundaries of an object. It will only identify the location of that edge and not the boundaries of the object; and

(d)    claim 1 of the Patent requires that the actual ‘boundaries’ (plural) of each shadowed region be determined, which means that the shape or outline of the obstruction must be determined.

Terms used in the Patent

187    Mr Appleyard said:

(a)    ‘image information’ cannot be just the output of a single sensor. A single output does not provide information about an image, it provides information about a single point. This is not sufficient by itself to constitute ‘image information’. There needs to be sufficient information from enough sensors such that shape of the obstruction can be determined;

(b)    ‘recognise’ is unclear. If it simply means ‘detect’ then there is no need to use the word ‘recognise’. If the word ‘recognise’ is used to describe the process by which ‘allowable obstructions’ are distinguished from ‘non-allowable obstructions’ (as described above) then Mr Appleyard can see why it would be used. In such a context it means more than just detection and connotes a state of affairs involving both known and unknown shadowed regions;

(c)    a ‘region’ cannot be a single point or even a number of single points. It is an area. In the context of the Patent, there needs to be sufficient information to determine the shape or boundaries of the region; and

(d)    based on the wording in the Patent of the term ‘obstruction’, it is something which is in the path of the blade (and not be the blade itself) which should not be there and which causes the safety system to trigger to prevent injury. The workpiece, unless bent in such a way as to have part of it bending upwards so as to be detected, is not an obstruction. Similarly, the anvil is not an obstruction.

3.    CONSTRUCTION OF THE PATENT

3.1    Principles

188    The claims of a patent serve to delineate the monopoly provided for by the patent: see Welch Perrin & Co Pty Ltd v Worrel (1961) 106 CLR 588 per Dixon CJ, Kitto and Windeyer JJ (at 609). In Electric & Musical Industries Ltd v Lissen Ltd (1938) 56 RPC 23 (at 39), Lord Russell said:

The function of the claims is to define clearly and with precision the monopoly claimed, so that others may know the exact boundaries of the area within which they will be trespassers.

189    The Court should approach the task of patent construction with a generous measure of common sense: Ranbaxy Laboratories Ltd v AstraZeneca AB (2013) 101 IPR 11 per Middleton J (at [108]); Beach J made a similar observation in GlaxoSmithKline Consumer Healthcare Investments (Ireland) (No 2) Ltd v Apotex Pty Ltd (2016) 119 IPR 1 (at [270]). The relevant principles are:

(a)    the proper construction of a claim is a matter of law: Decor Corporation Pty Ltd v Dart Industries Inc (1988) 13 IPR 385 per Sheppard J (at 400);

(b)    a claim is not to be read in the abstract, but is to be construed in the light of the common general knowledge and the art before the priority date: Welch Perrin per Dixon CJ, Kitto and Windeyer JJ (at 610); Kimberly-Clark Australia Pty Ltd v Arico Trading International Pty Ltd (2001) 207 CLR 1 per Gleeson CJ, McHugh, Gummow, Hayne and Callinan JJ (at [24]); Artcraft Urban Group Pty Ltd v Streetworx Pty Ltd (2016) 117 IPR 210 per Greenwood J (with whom Rares J concurred) (at [74]); GlaxoSmithKline (at [274]);

(c)    the words used in a specification are to be given the meaning that the normal person skilled in the art would attach to them, having regard to his own general knowledge and to what is disclosed in the body of the specification: Decor Corporation per Lockhart J (at 391) and GlaxoSmithKline (at [276]);

(d)    while the claims are to be construed in the context of the specification as a whole, it is not legitimate to narrow or expand the boundaries of monopoly as fixed by the words of a claim by adding to those words glosses drawn from other parts of the specification, although terms in the claim which are unclear may be defined by reference to the body of the specification: Welch Perrin (at 610); Kimberly-Clark (at [15]); GlaxoSmithKline (at [271]);

(e)    experts can give evidence on the meaning which those skilled in the art would give to technical or scientific terms and phrases and on unusual or special meanings to be given by skilled addressees to words which might otherwise bear their ordinary meaning: Sartas No 1 Pty Ltd v Koukourou & Partners Pty Ltd (1994) 30 IPR 479 per Gummow J (at 485-486), citing Glaverbel SA v British Coal Corp [1994] RPC 443 per Mummery J (at 485-486);

(f)    the Court is to place itself in the position of a person acquainted with the surrounding circumstances as to the state of the art and manufacture at the time: British Dynamite Co v Krebs (1896) 13 RPC 190 per Lord Chancellor Cairns (at 192); Kimberly-Clark (at [24]); Artcraft Urban Group (at [74]); GlaxoSmithKline (at [274]); and

(g)    issues of construction must be considered without any regard for the alleged infringing articles: Fresenius Medical Care Australia Pty Ltd v Gambro Pty Ltd (2005) 224 ALR 168 per Wilcox, Branson and Bennett JJ (at [95]); Australian Mud Company Pty Ltd v Coretell Pty Ltd (No 4) [2015] FCA 1372 per McKerracher J (at [114]). A court should construe the claims as ‘if the infringer had never been born’: Welcome Real-Time SA v Catuity Inc (2001) 51 IPR 327 per Heerey J (at [21]).

190    Because a patent is not a document inter partes, but a public instrument, this imposes limits on its construction: Welch Perrin (at 610); Kimberly-Clark (at [34]); Decor Corporation per Sheppard J (at 400); Pfizer Overseas Pharmaceuticals v Eli Lilly & Co (2005) 225 ALR 416 per French and Lindgren JJ (with whom Crennan J concurred) (at [249]). The Court is not free to amend the specification under the pretext of construction. Amendment is dealt with under a separate statutory mechanism. The hypothetical construct of the skilled addressee cannot be taken so far as to re-write or amend a claim of the specification: GlaxoSmithKline (at [12] and [14]). The Court is obliged to construe the claim as it is, rather than what it should have been: GlaxoSmithKline (at [12] and [14]).

191    The person skilled in the art is the hypothetical person to whom the patent specification is addressed (the skilled addressee or worker): General Tire & Rubber Co v The Firestone Tyre & Rubber Co Ltd (1971) 1A IPR 121 per Sachs, Buckley and Orr LJJ (at 134). As Finkelstein J observed in Root Quality Pty Ltd v Root Control Technologies Pty Ltd (2000) 177 ALR 231 (at [70]), the identity of the skilled addressee is relevant for a variety of purposes:

He is the person to whom the patent is addressed and who must construe it. He is the person whose knowledge will determine whether a patent is novel. He is the person who will judge whether a patent is obvious.

3.2    Terms in issue

192    With respect to claim 1, the terms in issue are:

(a)    ‘safety system’;

(b)    ‘image information’;

(c)    ‘recognise the presence of’;

(d)    ‘obstruction’;

(e)    ‘sufficient image information to determine the boundaries of the or each shadowed region’.

193    Those terms in integer 1.5 and integer 1.6 are required to be construed in context.

3.3    Applicants’ position on construction

194    Examining the five terms in issue sequentially, the applicants contend as follows.

195    ‘safety system’: the applicants contend that the words ‘safety system’ include any modality that provides a safety function (within a system which is for use with a machine having a moving part, as set out in the preamble to claim 1). This does not limit the safety function to stopping an operator having part of their body caught beneath the blade, or require the modality to serve solely a safety function. All that is required, the applicants say, is that the modality improves or facilitates the safety of an operator.

196    The applicants say the respondent’s approach to this term seeks to read down the term ‘safety system’ by reference to p 1 lines 5-7 of the Patent. In doing so, the applicants contend that the respondent seeks to limit the otherwise plain meaning of the words to a system solely concerned with one form of safety (i.e. an operator having a part of his/her body caught beneath the blade of a press brake as the blade descends). The applicants say the respondent has ignored p 11 lines 14-18, which refers to analysing the bending of work and considering the movement of shadows. This passage concerns different forms of safety to that identified on p 1 lines 5-7 of the Patent. This passage of the Patent identifies reducing the risk of a finger becoming trapped between the workpiece and the tool (i.e while the workpiece is being bent), as well as more generally referring to determining an ‘unsafe state’ during the bending process.

197    ‘image information’: the applicants assert that this is a broad concept, properly understood as including the information that is outputted from the light receiving means. This information is in turn provided to the processing and control means. ‘Image information’ could be, for example, the output from the sensors of a CCD or a camera. The image information in such an example would be the voltages produced by each diode/pixel in the CCD or camera. The applicants note that under cross-examination Mr Appleyard conceded that the term ‘image information’ was used in one of his own company’s patent application (for the Lazer Safe LZS-003 System high speed product) to describe the information output from a light receiver. Similarly, under cross-examination Mr Berry agreed that the term ‘image information’ as it appears in claim 1 of the Patent is the output of the light receiver.

198    In light of the evidence at trial the applicants submit that their construction of this term is to be preferred.

199    ‘recognise the presence of’: the applicants argue that this term must be properly understood in light of the phrase in which it appears, i.e. ‘recognise the presence of one or more shadowed regions’. This, the applicants say, requires no more than the ability to identify the existence of a shadowed region. As such, the applicants contend the term has the same meaning as ‘detect’.

200    The Macquarie Dictionary relevantly defines ‘recognise’ as:

1.    to know again; perceive to be identical with something previously known (he had changed so much that one could scarcely recognise him).

2.    to identify from knowledge of appearance or character.

3.    to perceive as existing or true; realise (to be the first to recognise a fact).

4.    to acknowledge formally as existing or as entitled to consideration (one government recognises another).

5.    to acknowledge or accept formally as being something stated (to recognise a government as a belligerent).

6.    to acknowledge or treat as valid (to recognise a claim).

7.    to acknowledge acquaintance with (a person, etc.) as by a salute.

8.    to show appreciation of (kindness, service, merit, etc.) as by some reward or tribute.

201    The Macquarie Dictionary relevantly defines ‘detect’ as:

1.    to discover or notice a fact, a process, or an action: to detect someone in a dishonest act.

2.    to find out the action or character of: to detect a hypocrite.

3.    to discover the presence, existence, or fact of.

…

202    Mr Acheson confirmed in his oral evidence that ‘recognise’ means the same as ‘detect’. Mr Acheson also stated that to recognise required a comparison. This need not be a comparison between images. As stated in Acheson # 1, ‘recognise the presence’ of a shadowed region requires prior knowledge of what will occur when a shadowed region is present. This could be, for example, the voltage output from a sensor dropping below a threshold. The relevant ‘comparison’ may therefore be the current output of a sensor or sensors against a pre-determined threshold. In his oral evidence, Mr Berry also confirmed that his use of the term ‘recognise’ in relation to the presence of an obstruction could be substituted for the word ‘detect’.

203    To the extent that Mr Appleyard gave the opinion that the term ‘recognise’ requires either the recognition of shapes, or alternatively a comparison with pre-stored images, the applicants contend that this ignores the plain meaning of the word as it appears in claim 1. The applicants argue that this would seek to import a limitation based Mr Appleyard’s view of the effect, on construction, of what is disclosed by a preferred embodiment described in the Patent.

204    In light of the evidence at trial the applicants submit that their construction of this term is to be preferred.

205    ‘obstruction’: as to this term, the applicants contend that ‘obstruction’ as it appears in claim 1 refers to any object which blocks light from the light emitter, thereby creating a ‘shadowed region’ on the light receiving means. This view is supported by the affidavit evidence in Acheson # 1, which was confirmed by Mr Acheson during cross-examination. An ‘obstruction’ can therefore include the tool, the workpiece, the anvil, or a part of the operator’s body. The oral evidence given by Mr Acheson on this point was to the effect that the tool, workpiece or anvil can all be within the illuminated region (i.e. they can obstruct the light beam), and can all cast a shadow on the light receiver. Some of these, such as the tool and the anvil, are expected to be present, and their presence should not (and does not in operation) require the tool to stop.

206    The applicants say that in the absence of any way of distinguishing between ‘good’ or allowable shadows (such as that cast by the tool, anvil or workpiece) and ‘bad’ shadows (such as that cast by the finger or hand of an operator) all shadows are potential obstructions. In general terms, the preferred embodiment of the invention described in the Patent is concerned with providing a safety system that can distinguish between these ‘good’ shadows and ‘bad’ shadows. For example, the Patent (at p 8 lines 8-12) refers to ‘safe’ shadow maps where ‘no obstruction is detected which would require stopping or slowing of the press brake’. Such a shadow map includes the forward edge of the tool and the anvil. In this context, the applicants say, the proper understanding of an ‘obstruction’ is any object casting a shadow.

207    In short, the applicants say the respondent attempts to narrow the meaning of the term ‘obstruction’ (in a bid to limit construction of claim 1 to a particular type of ‘image comparison’ device) by relying, with undue emphasis, on p 1 lines 5-7 of the Patent. The applicants contend that this fails to have proper regard to the plain meaning of the words of the claim. Where resort to the body of the specification is made, the applicants submit that the respondent’s approach fixates on one passage only and ignores other parts of the specification.

208    ‘sufficient image information to determine the boundaries of the or each shadowed region’: Consistent with the evidence of Mr Acheson, the applicants contend that this phrase is to be construed in the following way:

(a)    it refers to the resolution or quality of information that is required;

(b)    the resolution or quality of information must be such that the boundaries of the or each shadowed region are capable of determination. There is no requirement that a complete boundary, that is around the entirety of a shadowed region, must be determined by the system;

(c)    determination of a boundary involves identification of a transition from ‘light’ to ‘dark’ pixels (referred to as finding a ‘gradient’). Where a gradient is found between two adjacent pixels, an ‘edge’ exists. An edge is part of a boundary, but an edge is not a boundary within the proper construction of that term in claim 1;

(d)    a boundary of a shadowed region is a line that demarcates the area of the shadow that is cast. As an area is two-dimensional, it must be possible to find gradients and determine the ‘boundary’ in two dimensions; and

(e)    therefore, the phrase requires that there be sufficient image information (i.e. enough pixels at a suitable resolution and spacing) to allow the gradients (i.e. transition from light to dark pixels) to be found around the perimeter of the shadowed region. If the shadowed region comprises only one ‘dark’ pixel, this requirement is met if there is sufficient image information to identify this transition from ‘light’ to ‘dark’ between that one darkened pixel and the surrounding pixels.

209    The applicants contend that Mr Acheson is the only witness who gave any proper consideration to the meaning of this term as it appears in claim 1.

210    As to Mr Appleyard (for the respondent), the applicants say:

(a)    Mr Appleyard’s evidence in Appleyard # 1 was initially to the effect that this term requires that the system determine the shape of the obstruction and that the system uses this information. He simultaneously maintained that the term ‘sufficient image information’ was ambiguous and unclear;

(b)    Mr Appleyard later, in Appleyard # 3, gave evidence to the effect that the processing and control means must determine the boundaries of the shadowed region, and that the ‘sufficient image information’ was itself the boundaries of the shadowed region. This assertion, the applicants say, is illogical. The plain words of the claim require sufficient image information to determine boundaries. The ‘sufficient image information’ cannot itself be the boundaries. Rather, the ‘sufficient image information’ must be information which is capable of being processed to determine boundaries;

(c)    to the extent that Mr Appleyard is taken to assert that this phrase requires that the system determine the shape of the obstruction, the meaning Mr Appleyard’s gives to this term is indistinguishable from the meaning he gives to ‘image information’. Mr Appleyard’s conflation of these terms was highlighted when he expressed the view in Appleyard # 3 that ‘image information’ requires ‘sufficient information from enough sensors such that shape of the obstruction can be determined’; and

(d)    this conflation clearly illustrates Mr Appleyard’s failure to give proper consideration to the actual words used in claim 1. It also indicates that he has approached the entire construction exercise from the perspective that claim 1 is to be limited to a particular embodiment described in the Patent, and has attempted to give meaning to the words in the claim in order to arrive at this conclusion.

211    As to the respondent’s suggestion to Mr Acheson in cross-examination that his construction would mean that the system would never have to determine a boundary, would never have to stop the tool and that such a system could not work as a safety system, Mr Acheson’s response was that the system still had to have the information available to it, and that just because it did not determine boundaries did not mean that it could not use this information to operate as a safety system. The applicants say that this approach is entirely consistent with the final words of integer 1.6 ‘…and control movement of the part dependent on said image information’. That is to say, the claim contemplates that the system will have available to it image information of a particular resolution or quality (i.e. which is sufficient to determine boundaries), and it will use and process that information so as to effect control over the moving part.

212    The applicants submit that:

(a)    the relevant processing may require some determination concerning the location of an obstruction and/or a determination of part of a boundary of a shadow cast by an obstruction. Providing the system with ‘sufficient image information to determine the boundaries of the or each shadowed region’ ensures that the system is capable in each case of carrying out the required processing;

(b)    the framework of the dependent claims further supports the view that integer 1.6 does not require the system to carry out determination of a complete boundary. Rather, this integer is concerned with ensuring that the quality of image information that is available to the system is appropriate for the types of processing that is specifically required in the dependent claims;

(c)    the dependent claims of the Patent provide different embodiments or examples of the processing of the image information which can be undertaken in order to control the moving part. By way of example:

(i)    claim 2: the processing and control means either slows or stops the movement of the part if the processing and control means determines the presence of an obstruction in a predetermined or calculated area of the region;

(ii)    claim 3: the processing and control means calculates the positions of the obstructions relative to the part or relative to each other and slows or stops the part dependent on the relative positions;

(d)    similarly, each of claims 4, 13 and 14 refer to a determination or calculation that is carried out by the processing and control means in order to control the movement of the part. None of these claims require determining the complete boundary of the shadowed region cast by an obstruction.

Integer 1.5: a processing and control means arranged to receive image information from the light receiving means and thereby recognise the presence of one or more shadowed regions on the light receiving means cast by obstructions in the region

213    Integer 1.5 requires that there is a ‘processing and control means’ which receives the output from the ‘light receiving means’ (i.e. the image information). As a result, the system is able to detect (i.e. determine the existence of) one or more shadowed regions that are cast on the light receiving means by any object that blocks the light coming from the light emitter.

214    Again, the applicants contend that properly construed, integer 1.5 does not require any ability to recognise shapes, or any comparison with pre-stored images. The ‘image information’ that is received can simply be the output of the light receiver. In the case of a CCD or camera, this will be the voltage output by each diode/pixel, wherein these voltages represent the intensity of light observed by that diode/pixel.  Upon receiving this information, the processing and control means must be able to determine whether or not a shadowed region is present. This can be satisfied, for example, by comparing the voltage output against a pre-determined threshold. A shadowed region will be recognised as being present where the voltage falls below that threshold.

215    The reference to ‘one or more’ shadowed regions (cast by obstructions) in integer 1.5 contemplates that multiple shadowed regions may be present. The applicants argue that this reinforces the notion that the tool, workpiece and anvil, as well as the hand or finger of an operator, can be an ‘obstruction’ casting a shadow, as one or more of these would be required in addition to the hand or finger of an operator in order to create these multiple shadows. This ability to recognise multiple shadowed regions is a pre-condition, the applicants contend, to the capability described in the Patent to distinguish between ‘good’ and ‘bad’ shadows.

Integer 1.6: wherein the illumination of the region is such that the processing and control means has sufficient image information to determine the boundaries of the or each shadowed region and control movement of the part dependent on said image information

216    The applicants say integer 1.6 directs attention to the quality of the image information that is made available to processing and control means. The image information must be ‘sufficient’ to determine the boundaries of the or each shadowed region.

217    Integer 1.6 also directs attention, as noted by Mr Acheson in Acheson # 2, to the illumination of the region, which must be such that the ‘sufficient image information’ is available. This requires the image receiver being illuminated with light of sufficient intensity that a distinction can be made between the output from ‘light’ pixels (i.e. illuminated) and ‘dark’ pixels (i.e. shadowed).

218    Finally, integer 1.6 requires that the image information is sufficient to control movement of the part. This means that the processing and control means must be able to control the moving part, for example, by stopping, starting, slowing, or allowing continued motion of the part, based upon the image information it receives.

219    Dependent claims: It is common ground amongst the parties that no separate construction issues arise in relation to the dependent claims.

3.4    Consideration

220    As noted in Professor Michael Pendleton, ‘Construe Widely and Face Invalidity – Construe Narrowly and Miss Infringements: The Dilemma of Interpreting Patent Specifications’ (2004) 11(3) Murdoch University Electronic Journal of Law, patentees often face a dilemma as to the proper construction of the claims of the Patent. The stricter the approach to construction adopted, the easier it will be for the patent to be held to possess the requisite novelty and inventive step. Applying the same approach to construction will give rise to corresponding difficulty in proving infringement – that is to show, for example, that a variant which does not definitively transgress the claims of the patent specification is an infringement. However, if one adopts a more expansive and liberal approach to construction, attendant difficulties arise in demonstrating that the patent possesses the novelty and inventive steps, yet it is easier to demonstrate infringement.

The skilled addressee?

221    The relevant art and field is safety systems for operators of industrial machinery, such as press brakes, as at June 2002 or 2003. The person skilled in the art is therefore relevantly a hypothetical manufacturer and designer of safety systems for press brakes as at June 2002 or 2003. This person is the skilled addressee for the purposes of construction, infringement and validity.

222    Such a skilled addressee would have extensive practical design knowledge and experience in the available safety systems for press brakes, and the market for such systems, particularly in Australia and Europe (where the market was centred at the time).

223    Of the witnesses in this case, as I have already indicated, I consider Mr Appleyard, not Mr Acheson, as representative of the skilled addressee and as such I prefer Mr Appleyard’s evidence, on behalf of the respondent, on claim construction and infringement.

Claim 1

Integer 1.1: A safety system for use with a machine having a moving part arranged to move through a known path of movement, the safety system characterised by comprising

224    The term ‘safety system’ means a system, as further set out in the claim, which is designed to protect operators of the machinery from injury while the machinery is in use. For a press brake, this use is usually while the blade is pressing downwards to engage a workpiece, until that workpiece is bent and the blade rises again. It means anything which functions in any way at any time to protect the operator. A ‘safety system’ is directed to the protection of the operator of the press brake from injury, caused by, for example, getting a finger trapped between the blade of the tool and the workpiece. As the respondent submits, it can be differentiated from a control or productivity system which is concerned with other aspects of use of the press brake (for example, determining the correct angle of bending the workpiece). The ‘moving part’ is usually the tool. It is not the workpiece.

Integer 1.2: at least one light emitting means arranged to emit light

225    A ‘light emitting means’, taken in context, is an arrangement from a device to emit a large area parallel beam of light, or the beams described as modifications in the Patent.

Integer 1.3: the axis of the emitted light being generally perpendicular to the path of movement of the part such that a region including at least a portion of said path is illuminated

226    The ‘axis’ is shown in fig 4 of the Patent contained in Annexure A; across the path of the downward moving press blade in a press brake in operation. The ‘illuminated region’ is also shown by fig 4: parallel light beam 24 illuminates region 20 in the downward path of forward edge 48 of tool 12 (the press brake):

Integer 1.4: at least one light receiving means arranged to receive light from the one or more of the light emitting means which has passed through said region

227    The ‘light receiving means’ is described as an aperture in a screen and an image detection device such as a CCD or a projection screen with a camera ‘to observe the projected image’: Patent p 6 lines 7-9.

Integer 1.5: a processing and control means arranged to receive image information from the light receiving means and thereby recognise the presence of one or more shadowed regions on the light receiving means cast by obstructions in the region

228    The ‘processing and control means’ is described as being in the form of ‘a software program residing on a digital signal processor, a computer or embedded into a microcomputer which receives input from the output of the light receiving means’: Patent p 7 lines 1-3.

229    The ‘image information’ is received from the light receiving means (such as a CCD or a projection screen), but it is information used to ‘recognise the presence of one or more shadowed regions’ on the CCD or projection screen. In my view, this is an important factor. As the respondent says, it does not mean any information whatsoever about an ‘image’. ‘Image information’ means that the shape of the obstruction must be capable of being determined and sufficient to be used by the system.

230    Equally on this topic, in my view, the use of the word ‘recognise’, in context, is important, and means that the system is not just detecting a shadowed region, but the system recognises, that is, compares the shadowed region detected with what is previously known – the shape of obstructions. This is consistent with the recitation in the Patent (at p 6 line 23 to p 7 line 14), particularly the comparison of ‘the current image’ with the ‘stored image to determine the presence of any shadows’ (at p 7 lines 9-11). The reference in integer 1.5 to the ‘one or more shadowed regions’ accords with the notion that the system can determine the shape of the obstruction.

231    The emphasis is on obstructions when the machinery such as press brakes is descending.

232    Moreover, by the terms of claim 1 itself, it is quite clear that the blade itself cannot be an obstruction. It is the ‘moving part’ from integer 1.1 and the ‘part’ from integer 1.6, the movement of which is controlled. As Mr Appleyard said on behalf of the respondent, it cannot be both an obstruction and the moving part obstructing itself. Obstructions are something which is in the way of the blade of the tool and which should not be there during the descent of the tool towards the workpiece. Neither the blade nor the workpiece can be obstructions.

Integer 1.6: wherein the illumination of the region is such that the processing and control means has sufficient image information to determine the boundaries of the or each shadowed region and control movement of the part dependent on said image information

233    I accept the respondent’s submission that the reference to the processing and control means having ‘sufficient image information to determine the boundaries of the, or each, shadowed region’ is consistent in a common sense way with the focus of the claim on shape/outline recognition.

234    Specifically, the reference to ‘boundaries’ reinforces that the shapes or outlines of the obstructions must be determined sufficiently to be used by the system to control the movement of the tool. This is confirmed by the recitation at p 6 line 22 to p 7 line 14 and by the references to the processing and control means creating ‘a total picture made up of the image information’, which is referred to as a ‘shadow map’ and which is compared to ‘known safe shadow maps’ (at p 8 lines 1-12). I also prefer the respondent’s evidence and submission, in context, that an ‘edge’ is not a boundary, although it might be part of a boundary; a boundary encloses a region or area, extending around the entire perimeter of an object.