CCL Secure Pty Ltd (formerly Innovia Security Pty Ltd) v Visual Physics, LLC
[2019] APO 14
•27 March 2019
IP AUSTRALIA
AUSTRALIAN PATENT OFFICE
CCL Secure Pty Ltd (formerly Innovia Security Pty Ltd) v Visual Physics, LLC [2019] APO 14
Patent Application: 2013251640
Title:Security device for projecting a collection of synthetic images
Patent Applicant: Visual Physics, LLC
Opponent: CCL Secure Pty Ltd (formerly Innovia Security Pty Ltd)
Delegate: Dr V. Z. Kolev
Decision Date: 27 March 2019
Hearing Date: 12 September 2018, in Canberra, further submissions filed on 19 September 2018
Catchwords: PATENTS – section 59 opposition to grant of a patent – novelty – inventive step – utility – clarity – support – clear enough and complete enough disclosure – regulation 5.23 – optical security device – collection of synthetic images – pixels – image layer – binary grid of distributed digital images – collection of focusing elements – regulation 5.23 not invoked – opposition unsuccessful on all grounds – costs awarded
Representation: Counsel for the applicant: Mr Anthony Franklin SC
Solicitor for the applicant: Ms Katrina Crooks of Shelston IP Lawyers Pty Ltd
Patent attorney for the applicant: Mr Greg Whitehead of Shelston IP Pty Ltd
Patent attorney for the opponent: Mr Adrian Crooks of Phillips Ormonde Fitzpatrick
IP AUSTRALIA
AUSTRALIAN PATENT OFFICE
Patent Application: 2013251640
Title:Security device for projecting a collection of synthetic images
Patent Applicant: Visual Physics, LLC
Date of Decision: 27 March 2019
DECISION
For the purposes of deciding the opposition, I will not consult under the provisions of regulation 5.23 the declaration by Mr Gregory John Whitehead, dated 27 August 2018 with exhibits GJW-1 to GJW-4.
The opposition is unsuccessful on all grounds. Subject to appeal, I direct that the patent application proceeds to grant.
I award costs, according to the amounts specified in Schedule 8, up to the date of filing the request to amend the specification of the application against Visual Physics, LLC.
I award costs, according to the amounts specified in Schedule 8, from the date of filing the request to amend the specification of the application against CCL Secure Pty Ltd (formerly Innovia Security Pty Ltd).
REASONS FOR DECISION
Throughout this decision, unless explicitly stated otherwise, any reference to an Act or a section, subsection, etc. of an Act refers to the Patents Act 1990 (the Act), and any reference to Regulations or a specific regulation refers to the Patents Regulations 1991 (the Regulations).
Background
The matter relates to patent application 2013251640 (the Application) in the name of Visual Physics, LLC (the Applicant). The Application is the national phase entry of the international application PCT/US2013/037978, published as WO 2013/163287. The Application was filed on 24 April 2013 and claims a priority date of 25 April 2012.
The Application was advertised as accepted on 22 September 2016. A notice of opposition to grant was filed on 22 December 2016 by CCL Secure Pty Ltd (formerly Innovia Security Pty Ltd) (the Opponent).
The statement of grounds and particulars (the SGP) was filed on 22 March 2017. The SGP mentioned nine documents referred to as D1 to D9. The evidence in support was completed on 22 June 2017.
On 13 September 2017, the Applicant filed proposed amendments under section 104 and requested a stay in the opposition proceedings in view of the proposed amendments. On 22 September 2017, I issued a response in which I found the Applicant’s explanations supporting the request for a stay insufficient. I invited the Applicant to file further, more specific explanations within 14 days. In addition, I reminded the Applicant that requesting an extension of time to file evidence in answer is also an option that may be appropriate, provided the statutory requirements of subregulation 5.9(2) could be satisfied. No further correspondence on the matter of the stay was received. The proposed amendments were later allowed unopposed on 16 January 2018.
The evidence in answer was filed on 25 September 2017. The evidence in reply was filed on 28 November 2017.
The “Opponent’s Outline of Submissions” (the Opponent’s Summary or OS) was filed on 29 August 2018. The “Applicant’s Outline of Written Submissions” (the Applicant’s Summary or AS) was filed on 05 September 2018.
During the hearing, substantial disagreements between the parties became evident with respect to the proper interpretation of the term “pixel” as used in the claims, and in particular in the context of claims 1 and 2. Given the importance of the issue, I gave an oral direction which I confirmed in my correspondence dated 13 September 2018:
“The direction was to the effect that:
·Each party has 7 (seven) days from the date of the hearing to provide further submissions with respect to the meaning of the term ‘pixel’ as used on several occasions in the claims, including whether the term in question is used with the same meaning or not in all occasions found in claim 1.
·The parties do not have to provide further submissions if they consider this unnecessary for their case.
I am prepared to consider extending the period of 7 (seven) days if a party so requests and provides supporting explanations.”
Following this direction, the parties filed on 19 September 2018 the “Opponent’s Further Submissions” (the Opponent’s Further Submissions or OFS) and the “Applicant’s Additional Submissions” (the Applicant’s Further Submissions or AFS).
10. Relevantly, the present opposition is with respect to the Application as amended by the amendments filed on 13 September 2017, and is based on the SGP as filed.
Applicable Law and Onus
11. On 15 April 2013, the Intellectual Property Laws Amendment (Raising the Bar) Act 2012 commenced which resulted in significant amendments to the Act and Regulations affecting, inter alia, the standard of proof required for an opposition to succeed. For patent applications filed on or after the above commencement date, subsection 60(3A) applies:
“If the Commissioner is satisfied, on the balance of probabilities, that a ground of opposition to the grant of the standard patent exists, the Commissioner may refuse the application.” (emphasis added)
12. The Application was filed on 24 April 2013, hence subsection 60(3A) applies to the instant opposition. In addition, the filing date of the Application being after 15 April 2013 also means that the Application was examined under the amended provisions of the Act and Regulations following the Raising the Bar Act and the same are also applicable to the present opposition proceedings.
13. It is well settled that the Opponent has the onus of establishing the facts supporting the grounds of opposition, and this applies even though the standard of proof is “the balance of probabilities”.
Grounds of Opposition and Evidence
Grounds of opposition
14. The SGP lists the following as grounds of opposition:
- Not a manner of manufacture – paragraph 18(1)(a);
- Lack of novelty – subparagraph 18(1)(b)(i);
- Lack of inventive step – subparagraph 18(1)(b)(ii);
- Not useful – paragraph 18(1)(c);
- Lack of clear enough and complete enough disclosure – paragraph 40(2)(a);
- The best method known to the applicant of performing the invention not disclosed – paragraph 40(2)(aa);
- Claims not clear – subsection 40(3);
- Claims not succinct – subsection 40(3); and
- Claims not supported by matter disclosed in the specification – subsection 40(3).
15. However, in the Opponent’s Summary, the grounds of opposition are reduced to the following:
·Claims not clear;
·Lack of novelty;
·Lack of inventive step;
·Not useful;
·Lack of clear enough and complete enough disclosure; and
·Claims not supported by matter disclosed in the specification.
Therefore, my considerations will be limited to these grounds of opposition.
Evidence on file
16. The evidence filed in the opposition consists of the following documents:
Evidence in support consisting of:
· A declaration by Mr Axel Lundvall (Lundvall-1) dated 21 June 2017 with exhibits AL-1 to AL-6.
Evidence in answer consisting of:
· A declaration by Mr Timothy Paul Merchant (Merchant) dated 24 September 2017 with exhibits TPM-1 to TPM-8.
Evidence in reply consisting of:
· A declaration by Mr Axel Lundvall (Lundvall-2) dated 27 November 2017 with exhibits AL-7 to AL-13.
17. On 27 August 2018, the Applicant filed for consideration under regulation 5.23, a declaration by Mr Gregory John Whitehead, Patent Attorney, dated 27 August 2018 with exhibits GJW-1 to GJW-4 (the New Evidence). The Applicant proposed “that the Delegate consider this request at the opposition hearing”.
18. Introducing new evidence under regulation 5.23 was discussed in Merial Limited v Bayer Intellectual Property GmbH [2015] APO 16:
“I conclude that a decision under regulation 5.23 must have regard to the nature of the information and whether the information is likely, if not certain, to change the outcome of the opposition in a significant way.” (at [24])
My decision with respect to regulation 5.23 will be based on this test.
The Specification of the Application
19. The technical field of the invention is described as:
“The present invention generally relates to a security device for projecting a collection of synthetic images, and to a computer-implemented method for producing such a security device.” (at [2])
20. From the outset, I must make the general observation that the security device, its operation, and its method for producing could have been described and claimed in a somewhat easier to interpret manner. While the underlying physics behind the described security device is relatively simple and straightforward, the vague and imprecise language used at times is not very helpful and it appears to me that this language is the main reason for at least some of the Opponent’s concerns.
21. The section “Background and Summary of the Invention” explains that:
“Micro-optic film materials projecting synthetic images generally comprise (a) a light-transmitting polymeric substrate, (b) an arrangement of micro-sized image icons located on or within the polymeric substrate, and (c) an arrangement of focusing elements (e.g., microlenses). The image icon and focusing element arrangements are configured such that when the arrangement of image icons is viewed through the arrangement of focusing elements, one or more synthetic images are projected. These projected images may show a number of different optical effects.” (at [3])
22. After mentioning several prior art documents in the field, the specification continues:
“These film materials may be used as security devices for authentication of banknotes, secure documents and products. For banknotes and secure documents, these materials are typically used in the form of a strip or thread and either partially embedded within the banknote or document, or applied to a surface thereof. For passports or other identification (ID) documents, these materials could be used as a full laminate or embedded as an anti-counterfeit feature in polycarbonate passports.
The prior art film materials described above, which are known as moiré magnifiers, generally start with a two-dimensional (2D) array of identical image icons. They may, however, also start with image icons that are altered or modulated in ways that provide different effects such as changing images or images that slowly rotate, etc. For three-dimensional (3D) effects, these film materials are constructed using a ‘bottom-up’ approach in the sense that the view of a static object from the perspective of each individual lens is calculated spatially from a model of a static 3D object, and the corresponding icon is generated from the collection of the lens’ views. Using this approach, each icon is calculated individually based on the static model of the 3D object. The approach has at least the following limitations:
(a) The finished synthetic image is only a static 3D object;
(b) The finished synthetic image will have a ‘snap’ in the field of view; and
(c) The finished synthetic image is limited to a palette of at most one color, and furthermore one tone of that color. It is a binary image, and does not have any shading or grayscale.
The term ‘snap’, which will be described in more detail below, constitutes a large discontinuity in what the observer sees when the observer moves outside of the device’s range (but within its field of view) and looks at the device.” (at [5]-[7], emphasis added)
23. In essence, it appears that a “snap” occurs when, on gradual changing the angle of view, at certain point, an observer will see an abrupt (usually unintended) change in the projected image.
24. Having explained the limitations of the prior art, the description emphasises the way in which the instant invention can address those limitations and lists the advantages achievable by the instant invention:
“The drawbacks noted above are addressed by exemplary embodiments of the present invention, which use a ‘top-down’ approach in the sense that each desired complete synthetic image as seen by an observer from all given viewpoints is defined, and then each of these individual images (which is different from spatial information) that correspond to the different viewpoints is processed and then used to define a part of what each lens sees. The sum total of all of these viewpoint images will ultimately define a significant portion of the image plane that would normally contain only icons (‘image layer’). This approach will allow for the following major improvements over the prior art:
(a) The finished synthetic image can be, among other things: a moving 3D object; a dynamic (morphing or transforming) 3D object; a dynamic design of curves, abstract designs, shapes, photographs, 3D objects and images;
(b) The finished synthetic image can be designed such that there is no ‘snap’ in the field of view;
(c) The finished synthetic images can include ‘halftone’ effects similar to grayscale dithering. Furthermore, this method will help enable the coordination of several layers to ultimately produce synthetic images that incorporate full color dynamic designs and 3D images; and
(d) The finished synthetic images do not have to be sourced from models of 3D objects. The synthetic images can originate from any type of digital image (e.g., photographs, drawings, mathematical plots and curves, etc.).” (at [8], emphasis added)
25. The working of the invention is best understood with reference to the drawings, the most illustrative of which are reproduced below. I note that the exemplary devices use lenses as focussing elements. The basic principle is illustrated with reference to figures 1 to 3.
26. Figure 1 illustrates graphically the meaning of the term “optical footprint” which in the description is defined as:
“By way of technical background, for exemplary embodiments of the inventive security device that employ microlenses, each lens is able to project the entire image contained within its so-called ‘optical footprint’. As best shown in FIG. 1, an individual lens’ optical footprint is defined as the collection of every point on the image layer(s) that can be focused upon by the lens.” (at [26])
27. However, I note that even though the lens is technically able to project the entire image contained within its optical footprint, this entire image can never be observed from a single spatial location as illustrated in figures 2 and 3:
“The perception of a stationary observer relative to a lens’ optical footprint is shown in FIG. 2, the stationary observer seeing the lens as though it were representing a very highly magnified version of a very tiny subset of its optical footprint (e.g., colorless subset ‘B’ or colored subset ‘A’). In particular, the observer does not see an entire image [contained within the respective optical footprint] in the lens, but instead sees those image portions located at the focal point of the lens [wherein the location of this focal point within the optical footprint depends on the observer’s viewing angle]. As best shown in FIG. 3, the subset and thus color that is observed depends upon the location [the angular position] of the observer and, consequently, which bundle of collimated light the observer’s eye is receiving. The act or process by which small portions of a lens’ optical footprint are observed as the location [the angular position] of an observer changes is called ‘sampling’.” (at [27], emphasis added)
28. The emphasised portion of the above quotation represents a definition which I consider very important for the proper claim interpretation. The explanation continues:
“In general, if the lenses are sufficiently small, and the observer is sufficiently far away, the security device of the present invention may be characterized by the following statement:
An observer, looking through the collection of lenses that focus on the image layer(s) from a static viewpoint, sees that each lens is sampling the same place in each of the lens’ respective optical footprints simultaneously.” (at [28], original emphasis)
29. Importantly, “[a]n observer is shown as being very far away from the lenses relative to the size of the lenses, which means that the angle from the normal of the lens plane to the observer is essentially the same for each lens” (at [29]). It appears to me that, although they are somewhat related, even more important than the size of the lenses is the size of the security device itself (i.e. the largest distance between any two lenses), as it is this size (together with the distance to the observer) that determines the spread of the angles from the normal of the lens plane to the observer for each lens, and it is this spread that needs to be small. It is also explained that:
“The image that an observer sees from a given angle is hereinafter referred to as the viewpoint image. The collection of all viewpoint images that can be seen by an observer from an image layer can be coordinated to form many effects, perceived objects, and movements that have advantages over the optical effects demonstrated by prior art micro-optic security devices.” (at [31], emphasis added)
30. The basic principle demonstrated by reference to figures 1 to 3 is employed in an exemplary device for projecting nine viewpoint images, i.e. the numerals “1” to “9”, as illustrated in figures 7 to 12 below (figure 10 is not reproduced).
31. Figure 7-1 shows one of the lenses in the exemplary device with its respective optical footprint. Figure 7-2 shows a rectangular array of 30 lenses (six lenses tall by five lenses wide) as used in the device. The same lens array together with the corresponding array of 30 overlapping optical footprints on the image layer is shown on figure 7-3. Figure 7-4 illustrates one of the lenses with its domain, where “the term ‘domain’ is defined by the subset of each optical footprint that lies closer to its respective lens than to any other lens” (at [36], emphasis added). In this example, the domain is a square within the optical footprint of the lens, and it is digitised by being divided into digitized domain pixels (DD-pixels) as shown. Figure 7-5 illustrates the lens array with the corresponding array of 30 digitised domains. The array of digitized domains (viewed from above) is shown in figure 7-6. It is further explained that:
“The fact that there are nine pixels in each digitized domain means that a total of nine images can be prescribed to each domain [and to the device]. Each of these nine images, which are referred to as viewpoint images, can be seen from a different viewing angle or viewpoint. In this working example, each viewpoint image is a binary image (i.e., black or white only). Where there are thirty lenses in the lens array and thirty digitized domains in the image layer in this working example, each viewpoint image will contain exactly thirty pixels [referred to by the specification as ‘viewpoint image pixels’ or ‘VI-pixels’].” (at [37], emphasis added)
32. Figure 8-1 is similar to figure 7-4, however it also shows that each of the nine DD-pixels in the digitised domain is given a unique address within the domain, i.e. (1,1); (1,2); …; (3,3). The nine viewpoint images (i.e. the numerals “1” to “9”) are each assigned to a corresponding one of the addresses of the nine DD-pixels in the digitised domain as demonstrated by figure 8-2, i.e. numeral “9” is assigned the address (1,1); numeral “8” is assigned the address (1,2); …; and numeral “1” is assigned the address (3,3). This figure also illustrates that each numeral or viewpoint image has 30 VI-pixels (six VI-pixels tall by five VI-pixels wide), hence each lens in the array will represent one VI-pixel of a viewpoint image. Figure 9 illustrates how the numeral “9”, which is assigned the address (1,1), is distributed to a raster grid on the image layer. While not explicitly mentioned in the text explaining this figure, the raster grid on the image layer appears to contain the DD-pixels of the array of digitised domains illustrated on figure 7-6. The numeral “9” has 30 VI-pixels arranged in the same way as the 30 digitised domains in the array, each VI-pixel having a corresponding digitised domain. Each VI-pixel of the numeral “9” is distributed to its corresponding digitised domain such that the VI-pixel’s colour is reflected by the colour of the DD-pixel (in this corresponding domain) having address (1,1), i.e. the address with which the numeral “9” is associated. For example, the top-left VI-pixel of the numeral “9” is white, hence the DD-pixel with address (1,1) (see figure 8-1 for its location) in the top-left digitised domain is also white. The top-right VI-pixel of the numeral “9” is black, hence the DD-pixel with address (1,1) in the top-right digitised domain is also black. The same applies to every VI-pixel in the numeral “9”.
33. Since each numeral or viewpoint image is associated with a different address of a DD-pixel, the viewpoint images distributed to the raster grid do not interfere with each other. This is illustrated in figure 11 where all nine viewpoint images are distributed to the raster grid, in the same way as it was explained with respect to the numeral “9”, thus creating the final image layer. Since each DD-pixel in the raster grid represents the colour of one VI-pixel of one of the viewpoint images, nine viewpoint images having 30 VI-pixels each require a raster grid containing (9×30) = 270 DD-pixels.
34. Figure 12-1 illustrates the device where the lens array of figure 7-2 is placed over the image layer of figure 11. By varying the angle of view, the device can project all nine viewpoint images. Two such projected viewpoint images (the numerals “1” and “9”) are represented on figures 12-2 and 12-3, respectively, where two observers look at the device from different angles. It is clear that the same optical effect can be observed by a single observer on tilting the device, thus changing the angle of view. The effect is due to the fact that for each given angle of view, each lens projects or samples only a small portion of its optical footprint, this portion being within one specific DD-pixel of the corresponding domain.
35. Immediately after describing the above example device, the description explains that:
“As will be readily appreciated by those skilled in the art, the resolution of the viewpoint images can be increased as the number of lenses increases due to the proportionate correspondence between the number of viewpoint image pixels and the number of lenses. Similarly, the number of viewpoint images can be increased if the number of viewpoint image pixels in the digitized domains increase.” (at [41], emphasis added)
In the above, I consider that the underlined term “viewpoint image pixels” should read “DD-pixels” as the number of DD-pixels in one digitised domain is equal to the number of viewpoint images that a device can project. In addition, in the example device, a lens array of 30 lenses was used to project viewpoint images having 30 VI-pixels each, hence the reason for the above mentioned “proportionate correspondence” between these two numbers (as opposed to simply being equal) is not immediately apparent. However, it is worth mentioning that the viewpoint images of this example were described as “binary image[s] (i.e., black or white only)”. For this type of images, indeed the number of VI-pixels is equal to the number of lenses. The described need for a “proportionate correspondence” becomes clearer when the viewpoint images are greyscale (halftone) viewpoint images.
36. This is illustrated on figure 15.
In this example, it is described how a cluster of four lenses could be used to represent each VI-pixel, thus allowing three levels of grey (25%, 50%, and 75%) in addition to black (100%) and white (0%) for each VI-pixel. It appears that this could only be done at the expense of the viewpoint image resolution (in terms of the number of VI-pixels in the viewpoint image). The number of VI-pixels in a greyscale viewpoint image produced in accordance with figure 15 will be only a quarter of the number of the lenses in the device, wherein (as mentioned earlier) for binary images, the number of VI-pixels is equal to the number of the lenses in the device.
37. In addition to providing a way to project greyscale images, the description also explains how the invention can be used to create other optical effects like 3D images, and image collections exhibiting “no snap”.
38. The specification ends with 20 claims. Claims 1 and 6 are the only independent claims and are reproduced below together with dependent claims 2 and 8 which, being related to the generation of greyscale viewpoint images, caused some of the Opponent’s concerns (some formatting added to assist readability):
“1. A security device for projecting a collection of synthetic images, which comprises:
a collection of focusing elements, with each focusing element having an optical footprint; and at least one image layer,
the collection of focusing elements and the at least one image layer together projecting a different image as the device is viewed at different angles,
wherein, the at least one image layer includes a binary grid of distributed digital images,
the binary grid comprising a plurality of pixels where each pixel in the binary grid is either on or off,
wherein, the binary grid of the at least one image layer is made up of an array of discrete digitized domains,
each domain constituting an identical or substantially identical subset of each focusing element’s optical footprint,
the domains being discrete in that no two subsets overlap and every point in each subset is closest to its respective focusing element,
each domain being divided into a number of discrete pixels equal to the number of images with each pixel assigned a location within the respective domain,
wherein, each image is processed digitally by sampling pixels from the binary grid,
the number of pixels in each digitally processed image being equal or proportionate to the total number of focusing elements,
the pixels in each digitally processed image being distributed to the same location within each digitized domain,
such that each location within one digitized domain is marked with the color of a pixel from a different digitally processed image,
allowing for the device to project a different image as the device is viewed at different angles.
2. The security device of claim 1,
which projects a collection of grayscale or halftone synthetic images, wherein each image is a grayscale image with a reduced color palette, and
wherein a cluster of focusing elements is used for each image pixel.
6. A computer-implemented method for producing a security device for projecting a collection of synthetic images, wherein the security device is made up of
a collection of focusing elements in the form of a focusing element sheet, each focusing element having an optical footprint; and at least one image layer,
the focusing elements and the at least one image layer together projecting a different viewpoint image as the device is viewed at different angles,
the method comprising:
(a) forming the at least one image layer by:
(i) compiling a collection of different raw viewpoint images with each raw viewpoint image prescribing what an observer should see when viewing the security device from a given angle;
(ii) choosing a domain for each focusing element in the focusing element sheet, and arranging the domains in the form of a grid on or within the at least one image layer,
the domains constituting identical subsets of each of the focusing element’s optical footprints such that no two subsets overlap and every point in each subset is closest to its respective focusing element once the at least one image layer is placed in a focal plane of the collection of focusing elements,
wherein exact registration between the domains and the focusing elements is not required;
(iii) digitizing each chosen domain by dividing each domain into a number of discrete pixels, which will each represent a portion of a different viewpoint image,
the number of pixels in each digitized domain being equal to the number of different viewpoint images,
the digitized domains forming a raster grid;
(iv) digitally processing each of the different raw viewpoint images to form binary images,
wherein the number of pixels in each digitally processed viewpoint image is equal to, or proportionate to, the total number of focusing elements in the focusing element sheet that will be used to represent the desired complete image;
(v) distributing the collection of different processed viewpoint images into the raster grid by marking each digitized domain pixel with the color of the corresponding viewpoint image pixel through a process called ‘distribution’, which involves
assigning an address to each pixel in each digitized domain, and then
assigning one image to each pixel having the same address in each digitized domain, such that each location within one digitized domain will be marked with the color of a pixel from a different processed viewpoint image; and
(b) placing the formed image layer or layers in a focal plane of the focusing element sheet.
8. The computer-implemented method of claim 6,
which produces a security device that projects a collection of grayscale or halftone synthetic images, wherein each raw viewpoint image is a raw grayscale or halftone viewpoint image,
wherein the raw viewpoint images are modified by reducing the number of shades of gray in each image’s color palette, optionally dithering the remaining shades of gray in each image’s color palette, and then
representing each such processed viewpoint image as a finished binary image,
wherein after distributing the collection of different finished binary images into the raster grid, a cluster of focusing elements is used for each viewpoint image-pixel.”
The Person Skilled in the Art and the Experts for the Parties
39. The hypothetical “person skilled in the art” is a well-established legal concept and I do not consider it necessary to discuss the relevant Authorities.
40. The Opponent submits that:
“The skilled addressee of AU’640 [the Application] would [sic] a person with a practical interest in the design and manufacture of security devices having optically variable effects caused by micro-optics and in particular for use on objects of value such as banknotes. Having regard to his education and experience, and familiarity with the common general knowledge, the opponent submits that Lundvall is plainly representative of the skilled addressee (Lundvall 1 [1] to [5] and [8] to [33]).
By comparison, Merchant displays a limited understanding of matters which were generally known in the field of security devices, particularly in relation to banknotes. For example, in paragraph [24] of Merchant it is said that with one exception, he was not aware of any products at the priority date which were capable of forming floating or 3-dimensional images. This displays a distinct lack of familiarity with the common general knowledge. The product Unison, produced by the company Nanoventions, and capable of producing such images was first disclosed to the document security industry in 2004. The Unison product was incorporated in the Swedish 1000 Krona bank note in 2006 and was widely discussed in a variety of publications thereafter. Any person skilled in the art would have been aware of the Nanovention’s Unison product by at least April 2012 (Lundvall 2 [4] to [9]). A commercial embodiment of the invention described in International Patent Application WO2010/094691 was also available in the market from 2009 and was widely publicised to the document security and packaging industries (Lundvall 2 [10] to [15]).
Having regard to Merchant’s lack of awareness of the common general knowledge, to the extent that the evidence reveals areas of dispute between the experts, the evidence of Lundvall should be preferred.” (OS at [10]-[12], emphasis added)
41. I find it interesting that the Opponent apparently attempts to determine whether a particular person is representative of the person skilled in the art based on their awareness or otherwise of what the Opponent considers to be common general knowledge. I consider that the inverse to this is the proper procedure, i.e. firstly identifying the persons who, based on their qualifications and experience, can be considered representatives of the person skilled in the art, and then establishing the common general knowledge with reference to the evidence provided by these persons.
42. Mr Merchant has over 35 years of experience in the security industry (Merchant at [1]). His involvement also includes optical security devices:
“In 1998, I became the Managing Director of OpSec Asia, Pte Ltd, part of the Optical Security Group, based in Singapore and Vietnam. Optical Security Group is a United States based business providing anti-counterfeiting technologies for governments and corporations worldwide. I was responsible for setting up the Asian business and managing its operations.” (Merchant at [7])
“In 2002, I established, and became President of, Intellectual Product Protection, Inc, a security consulting business with offices in Singapore, Thailand and the USA. As part of my role, I have advised Governments and corporations on ID and product security, including design of security devices and technology, materials compatibility and forensic analysis. In 2007 I also founded the Forensic Services Lab in support of investigations to identify sources of counterfeit manufacturing, particularly within the pharmaceutical industry. Our clients have included Microsoft (completing an adversarial analysis project on the ability to place images on the readable side of CD or DVD and maintain functionality, and evaluating a security device for implementation of all retail labels and Certificates of Authenticity for Microsoft), Pfizer, Johnson & Johnson, Cisco, the United States Department of State, Bank of Thailand (developing the specifications and technical design for the holographic stripe on the 100 bhat banknote in Thailand), the Passport Office of Bangladesh and First Research Institute Beijing, for which we assisted in the development of the e-Passport for China. During this time I also developed an optical security feature called ‘HaloSec’ for polycarbonate ID that lies ‘dormant’ until the issuance of the ID, when it is activated by laser, and a non-embossed security foil technology called ‘VariSec Foil’, used today in over 125 million ePassports around the world in various countries.” (Merchant at [8], emphasis added)
43. I note that Mr Merchant was also involved in inventive activities in the field:
“Have obtained patents for a latent image printing process and for a security fiber used in security paper and passports.
…
Developed a new infra-red security foil that is patent pending in the United States. Also developed a security feature for identification documents made from polycarbonate, which has been submitted for a patent.” (Mr Merchant’s Curriculum Vitae filed as exhibit TPM-1)
However, the same can be said for Mr Lundvall (see Lundvall-1 at [4] and [21]).
44. In conclusion, based on the evidence on file, I consider that both Mr Lundvall and Mr Merchant are qualified to give evidence in this opposition.
The New Evidence
45. The Applicant points out that:
“In support of the case on obviousness the Opponent relies upon the evidence of Lundvall, who is the inventor of D2. It is important to note that Lundvall has links to the Opponent and the Opponent has not provided any evidence from an independent expert.
For the Applicant, evidence relevant to the obviousness question is given by a witness who is totally independent – Merchant, who has an extensive experience in the security industry and in particular in the field of passport or other document security.” (AS at [98]-[99], emphasis added)
46. The Applicant emphasises the independence of Mr Merchant and the links between Mr Lundvall and the Opponent. In addition, as already mentioned, the Applicant filed the New Evidence which, according to them, “is highly relevant to the independence or otherwise of the Opponent’s expert witness, Mr Axel Lundvall, CEO and co-founder of Rolling Optics AB”.
47. It is worth noting that it is not uncommon for parties to opposition proceedings to file evidence by experts that are not completely independent. While the New Evidence could establish the existence of partnership between the Opponent and Mr Lundvall’s employer, as it will become apparent later in this decision, this fact is not something that “is likely, if not certain, to change the outcome of the opposition in a significant way”.
48. Therefore, I conclude that I have no reasons to introduce the New Evidence into the opposition under regulation 5.23, hence for the purposes of deciding the opposition, I will not consult the New Evidence under the provisions of this regulation.
Claim Interpretation and Clarity under Subsection 40(3)
49. In this decision, any reference to Macquarie Dictionary, unless explicitly stated otherwise, refers to the online edition as reviewed on 14 September 2018.
50. For the purpose of this decision, I will use the word “colour” to denote the visual attribute that can be used to distinguish two neighbouring uniformly coloured areas from each other. In other words, if two areas with a common boundary have different colours then the boundary will be visible, whereas if the two areas have the same colour, the boundary will not be visible and the two areas will merge visually into one area. In that regard, black and white are also considered colours and different hues, shades or nuances of the same colour (e.g. dark green and light green) are considered different colours as long as they are visually distinguishable.
The meaning of the term “pixel”
51. The proper interpretation of the term “pixel” in context is very important for the outcome of my decision. Each one of the independent claims 1 and 6 uses this term in eight instances. Claims 2 and 8 are appended to claims 1 and 6, respectively, and use the terms “image pixel” and “viewpoint image-pixel”, respectively. It is important to ascertain the exact meaning of the term “pixel” in each instance, i.e. the exact physical or abstract object it refers to. As I already mentioned, I invited and the parties provided further submissions on the matter.
52. According to the Opponent’s Further Submissions, claim 1 refers to three categories of pixels: “grid pixels”, “domain pixels”, and “image pixels”. The Opponent asserts that nothing in the specification indicates that the “grid pixels” are the same as or equivalent to the “domain pixels”, and following some calculations with respect to the three categories of pixels, concludes:
“The applicant submits that each of the eight references to the term ‘pixel’ in claim 1 is a reference to the same thing. The opponent submits that this is not the correct construction of claim 1.” (OFS at [13])
“There is no defined relationship between the grid pixels and the domain pixels other than they both exist as part of the image layer.” (OFS at [16], emphasis added)
“Because the number of image pixels into which each digitally processed image is divided is equal or proportionate to the total number of focusing elements, the number of image pixels is essentially unrestricted. The number of image pixels may correspond to the number of domain pixels, the number of grid pixels or neither.” (OFS at [17], original emphasis)
53. Having considered the relevant legal principles, the Applicant states that:
“As explained below ‘pixel’ has the same meaning wherever it is used in claim 1. However ‘image pixel’ in claim 2 has a slightly different meaning.” (AFS at [9])
“The Applicant accepts the Opponent’s general contention that the word ‘pixel’ in ordinary language refers to an element of a picture. However it is wrong to suggest that, in the context of the Application and in particular as used in the claims of the Opposed Application, the nature or identity of the ‘pixels’ changes. As is demonstrated below, the term is used consistently throughout the description and the claims to refer to the particular element of an image assigned to a place in the binary grid of which the image layer of the claimed device is formed.” (AFS at [10])
54. The Applicant asserts that “[a]ll eight references in claim 1 to ‘pixels’ are to these DD-pixels” (AFS at [13]), and explains:
“It is important to note that, in the phrase ‘the number of pixels in each digitally processed image being equal or proportionate to the total number of focusing elements’ the words that we have emphasised in bold are not referring to the total number of pixels in each domain, but the number of pixels in the image viewed from a particular angle. For example, in the embodiments described and illustrated with reference to Figures 7-12 each domain is divided into nine pixels and there are nine different images. But, when a single image (for example the numeral 9 – see Figure 9) is viewed. each lens will focus on one pixel in each domain with the ‘on’ pixels generating the image (in this case the numeral 9). Thus the number of pixels in each viewpoint image is equal to the number of focusing elements.” (AFS at [15], original emphasis)
“the number of pixels in each digitally processed image being equal or proportionate to the total number of focusing elements – for the reasons explained above the number of pixels in each image (not domain) is equal to the number of focussing elements since one pixel in each image is distributed to or seen by each lens.” (AFS at [18], original emphasis)
55. While I find the parties’ further submissions helpful, I am still unsure that I can entirely agree with either of them. For example, I am not convinced that without giving an extremely broad interpretation to the term “pixel”, one can conclude that the “grid pixels” and the “domain pixels” could be different despite “both exist[ing] as part of the image layer”. It is also unclear to me how the Applicant’s interpretation explains the presence of the phrase “being equal or proportionate” in claim 1.
56. I will start my interpretation of the term “pixel” by referring to Macquarie Dictionary:
“noun 1. any of the extremely small discrete elements, known as dots, which together make a graphic image as on a television or computer screen or as produced by a digital camera. Each pixel consists of red, green, and blue components which together represent a particular colour to the human eye. The more pixels there are in an image, the greater its resolution.
…
[from pix picture + EL(EMENT)]”
57. It does not appear that Mr Merchant has provided a definition of the term in question, whereas Mr Lundvall states that:
“A pixel is simply the smallest definable element in a raster image. By definition, any such digital image is composed of pixels.” (Lundvall-2 at [20])
58. Neither of the Experts has provided a definition for the term “raster image”. Macquarie Dictionary also does not appear to provide a specific definition of the term “raster image”, however I consider the following definition of the word “raster” as most relevant:
“2. Computers a matrix of cells or pixels fixed in a grid of rows and columns, with each cell containing a value representing information, as colour, location, temperature, etc.”
59. From the above discussion, I can conclude that any raster image is an image produced by the above “matrix of cells or pixels fixed in a grid of rows and columns”, and that pixels always belong to raster images being their smallest definable elements. It follows that in order to interpret the meaning of the term “pixel” as used in various instances in the claims, I need to identify the raster image(s) to which the pixels mentioned in each instance belong. In the context of the claimed invention, I consider that there are two distinct raster images (or types of images) formed by pixels. The first is any one of the different synthetic images that are projected by the security device as the device is viewed at different angles. For brevity, I will refer to each image of this type as a synthetic image (examples of such images are presented on figures 12-2 and 12-3). The second image is the image formed on the image layer of the device. While I note that this image may not represent any recognisable physical or abstract object and it is not intended to be viewed by an observer through the operation of the security device, nonetheless it is a raster image printed or otherwise formed on or in the image layer. For brevity, I will refer to this image as the layer image (example of this image can be seen on figure 11).
60. In addition, since pixels are the smallest definable elements in a raster image, I consider that each pixel must be uniform within the limitations of the applicable production technology. In other words, a pixel should not have any irregularities or non-uniformities that are deliberately included in it by design. Any element in a raster image, the element containing optically distinguishable structure(s) on a scale comparable to the size of the element, would not be the smallest definable element of the raster image since smaller elements within this element would be definable. In that respect, I consider a pixel in the present context to be different from a screen pixel, which (as explained by Macquarie Dictionary) has three colour components (i.e. red, green, and blue) to create the colour of the pixel. For example, if one half of an element in a raster image is black and the other half is white, I do not consider that this would represent a grey pixel. Instead, this would mean that this element cannot be considered a pixel of this raster image as it can be split into two smaller elements (one black and one white). However, I do not consider that irregularities within any pixel on a scale much smaller than the size of the pixel (e.g. dots of ink visible under much higher magnification than that provided by the security device of the instant invention) would prevent it from being properly considered a pixel.
Interpretation of claim 1
61. Claim 1 defines that:
“the … image layer includes a binary grid of distributed digital images,
the binary grid comprising a plurality of pixels
where each pixel in the binary grid is either on or off”.
It is clear that these binary grid pixels form the image of the image layer, hence they are pixels of the layer image (or layer image pixels).
62. Claim 1 also defines that:
“the binary grid of the … image layer is made up of an array of discrete digitized domains,
each domain constituting an identical or substantially identical subset of each focusing element’s optical footprint, the domains being discrete in that no two subsets overlap and every point in each subset is closest to its respective focusing element,
each domain being divided into a number of discrete pixels equal to the number of images with each pixel assigned a location within the respective domain”.
Since the domains make up the binary grid of the image layer, I consider that each domain is (or represents) a portion of the image layer located in the optical footprint of one focusing element, such that the points in this portion are projected by this focusing element. Hence each one of the pixels into which each domain is divided (i.e. the domain pixels) is also a layer image pixel.
63. A question arises as to whether the domain pixels and the grid pixels refer to the same objects. Adopting Mr Lundvall’s interpretation that the pixel is “the smallest definable element in a raster image”, and given that the domain pixels and the grid pixels are pixels of the same raster image (i.e. the layer image), I am not convinced that they could refer to different objects as stated by the Opponent:
“There is nothing in the specification or in the language of claim 1 which indicates that the grid pixels of the binary grid are the same as or equivalent to the domain pixels of the digitised domains. For example, nothing in claim 1 precludes and [sic] image layer having:
i) a binary grid in the form of a 100×100 grid of grid pixels; and
ii) a 10×10 array of digitised domains, with each domain being a 5×5 array of domain pixels.
In such an arrangement, the image layer consists of 10,000 grid pixels and 2,500 domain pixels, with each domain pixel represented by a 2×2 array of grid pixels.” (OFS at [6]-[7], emphasis added)
Indeed, if “each domain pixel [is] represented by a 2×2 array of grid pixels”, then the domain pixels will not be the smallest definable elements of the layer image, hence they will not be layer image pixels. Even if I assume that every domain is a separate raster image having separate pixels, i.e. the domain pixels, wherein each domain pixel is represented by a 2×2 array of grid pixels, then the domain pixels will not be the smallest definable elements of the domain – the grid pixels forming the domain pixels will be smaller. I do not consider that the evidence supports the view that a pixel is a somewhat arbitrarily defined portion of a raster image. On purposive construction, I conclude that both the grid pixels and the domain pixels refer to the same objects, i.e. the layer image pixels.
64. Another important consideration is the meaning of the expression “each pixel in the binary grid is either on or off”. I was unable to find a clear definition in the context of the invention of the term “binary” as well as of the pixels being “either on or off”, neither in the specification, nor in the expert evidence. The description provides the following explanations, however due to the language used, I do not consider them to be definitions:
“In contrast to a moiré magnifier, which has an array of more or less ‘continuous’ images, an exemplary embodiment of the image layer(s) in the security device of the present invention is a binary grid of distributed digital images where each pixel in the grid is either ‘on or off’ (i.e., colored or uncolored).” (at [11], emphasis added)
“In the previous section, simple viewpoint images that were binary in the sense that the viewpoint-pixels were either black or white were distributed into the image layer(s).” (at [42], emphasis added)
65. In interpreting the feature in question, Mr Merchant refers to the body of the specification:
“I have also considered claim 1 of the proposed claims. This adds the feature to claim 1 that the image layer includes a binary grid of distributed digital images, the binary grid comprising a plurality of pixels where each pixel in the binary grid is either on or off. I have described this feature of the invention of the Opposed Application above at paragraph 51.” (Merchant at [80])
I consider that the above reference to paragraph [51] was supposed to refer to the paragraph quoted below:
“At paragraph [0011] it is noted that in contrast to a moiré magnifier, which has an array of more or less ‘continuous’ images, the image layer in the security device of the Opposed Application is a binary grid of distributed digital images where each pixel in the grid is either ‘on or off’ (i.e coloured or uncoloured). This can be seen, for example, in Fig 8, frame 2 where each numeral is formed as a combination of black and white square pixels. This is a particular feature of this document as compared to all the other documents I have considered which all provide ‘continuous’ digital images rather than images formed from discrete pixels.” (Merchant at [66]).
66. Mr Lundvall comments on these features on several occasions:
“Almost any two colour image layer produced for the purpose of creating a moiré magnification device will consist of a binary grid of distributed digital images. Any such image may be transformed into an electronic digital image and will therefore inherently consist of a binary grid of pixels which are either on or off. Further, many of the methods used to actually print such an image layer will require the image to be produced as a binary grid of pixels which are either on or off. The larger the pixels, the lower the resolution of the resulting image. Even though the images described in AU’640 [the Application] have lower resolution than is typical for moiré magnification devices, this difference in scale or resolution is only a matter of degree. The synthetic images, no matter what their resolution, are produced by the same fundamental design process.” (Lundvall-2 at [19], emphasis added).
“In paragraph [80] of Merchant there is a discussion of the amended form of claim 1 of AU’640 [the Application], in particular the addition of the requirement that;
the at least one image layer includes a binary grid of distributed digital images, the binary grid comprising a plurality of pixels where each pixel in the binary grid is either on or off
For the reasons discussed above, any two colour (e.g. black and white) digital image meets this description. Other than limiting the pixels to being in one of two states, I do not consider that this feature adds any limitation to the claim which already defined the image layer as being made up of an array of digitized domains divided into discrete pixels.” (Lundvall-2 at [24], emphasis added)
“As discussed above, the requirement that each pixel be either on or off simply defines a two colour (e.g. black and white) image. It is self-evident to me that the images formed as part of the integral image device described in AUS’366 could be two colour (e.g. black and white) images.” (Lundvall-2 at [25], emphasis added)
“In paragraph [91] of Merchant it is said WO’691 does not disclose the above mentioned binary pixel feature. For the reasons set out in my First Declaration I disagree. In describing the most basic embodiment of the invention of WO’691 by reference to Figure 2A it is said that part of the image cell contains optically distinguishable structures while another part is ‘empty’. It is clear that this is describing a binary arrangement in which pixels are either ‘on’ (optically distinguishable) or ‘off’ (empty).” (Lundvall-2 at [27], emphasis added)
“In paragraph [110] of Merchant, it is again said that WO’691 does not disclose the formation of a binary image and instead teaches the use of optically distinguishable structures within cells. For the reasons set out above, I do not consider there to be a difference between pixels which are on or off and pixels with a presence or absence of optically distinguishable structures. These are simply two different ways of describing the same thing.” (Lundvall-2 at [36], emphasis added)
67. Analysing Mr Lundvall’s evidence, it appears to me that his general comment that the expression in question is only “limiting the pixels to being in one of two states” is made in the context of two-colour images (e.g. using the colours black and white), or the presence or absence of optically distinguishable structures representing “on” and “off”, respectively. A more general consideration of the pixels being in one of two states is not provided, therefore I consider that the two possible states are limited to the presence or absence of a specific colour or optically distinguishable structures in the pixel.
68. In addition, claim 1 defines that “each location within one digitized domain is marked with the color of a pixel from a different digitally processed image”. Given that “each [domain] pixel [is] assigned a location within the respective domain”, and that domain pixels are layer image pixels, on purposive construction, a location within a digitised domain can be marked with the colour of a pixel from a different image only if the layer image pixel that is assigned this location within the domain is marked (or has) the same colour as that pixel from a different image. If follows that, on proper construction, the expression “each pixel in the binary grid is either on or off” defines that each layer image pixel of the security device must be uniformly coloured in either one of two possible colours.
69. Claim 1 further defines that (the features are labelled for easier reference):
“(i) each image is processed digitally by sampling pixels from the binary grid,
(ii) the number of pixels in each digitally processed image being equal or proportionate to the total number of focusing elements,
(iii) the pixels in each digitally processed image being distributed to the same location within each digitized domain,
(iv) such that each location within one digitized domain is marked with the color of a pixel from a different digitally processed image”.
This part of the claim is an object of serious disagreements between the parties and I will consider each of the above four features in the context of the claim.
70. Feature (i) defines that “each image is processed digitally by sampling pixels from the binary grid”. As mentioned earlier in this decision, the term “sampling” is defined in the specification:
“The act or process by which small portions of a lens’ optical footprint are observed as the location [i.e. the angular position] of an observer changes is called ‘sampling’.” (at [27], emphasis added)
No alternative meaning for this term is proposed and the above definition makes it clear that the sampling occurs during the projection of the synthetic images by the device. Hence the word “image” must refer to the synthetic image viewed at a particular angle, and it is this synthetic image that “is processed digitally by sampling pixels from the binary grid”.
71. As the binary grid contains layer image pixels, “sampling pixels from the binary grid” during the projection of an image should be interpreted as referring to the layer image pixels because it is the pixels of the layer image that are sampled by the focusing elements to project the synthetic image. Therefore, I consider that feature (i) should be interpreted as defining that each synthetic image is created digitally by the process of sampling layer image pixels from the binary grid.
72. Since the synthetic image is created digitally by the process of sampling (the claim even uses the phrase “processed digitally”), it is then a digitally processed image. I note that the term “digitally processed image” is used in all three of features (ii), (iii), and (iv). I consider it reasonable to proceed on the basis that the term in question refers to the same object on all occasions, and that this object is the synthetic image. This should be reconsidered if the following attempt to interpret features (ii), (iii), and (iv) on this basis results in irreconcilable contradictions.
73. Feature (ii) defines: “the number of pixels in each digitally processed image being equal or proportionate to the total number of focusing elements”. As discussed above, the term “digitally processed image” refers to the synthetic image, hence “pixels in each digitally processed image” refer to the pixels of the synthetic image (or synthetic image pixels) as distinct from the layer image pixels. In that respect, although the term “pixel” itself has the same general meaning as discussed earlier in this decision, the synthetic image pixels and the layer image pixels are pixels of different raster images, hence they may not necessarily refer to the same objects.
74. Macquarie Dictionary defines the word “proportionate” as: “proportioned; being in due proportion; proportional” (emphasis added). In addition, the most relevant definition of the word “proportional” (again in Macquarie Dictionary) is: “Mathematics having the same or a constant ratio” (emphasis added).
75. Since the synthetic image pixels are generated through the sampling process, prima facie, it would appear that the number of these pixels should be equal to the total number of focusing elements. Indeed, each focusing element samples a different layer image pixel which is uniformly coloured in either one of two possible colours, hence each focusing element projects a single colour. Therefore, it is impossible to have more synthetic image pixels than focusing elements. It is, however, possible that a synthetic image may be designed such that several neighbouring focusing elements project the same colour thus forming a group. If the collection of focusing elements is divided into such groups, then each group can be considered to project a single synthetic image pixel, hence the number of pixels in a synthetic image can be smaller than the total number of focusing elements. Therefore, on purposive construction, I conclude that feature (ii) defines that the ratio of the total number of focusing elements divided by the number of pixels in each synthetic image must be a constant greater than or equal to one.
76. Feature (iii) defines: “the pixels in each digitally processed image being distributed to the same location within each digitized domain”. I note that the grammatical construct “being + past participle” is used on several occasions in the claim (e.g. “each domain being divided”, “the number of pixels … being equal or proportionate to”, “the pixels … being distributed”). In my view, this grammatical construct can be used in present progressive and past progressive passive forms, hence its use is not limited to expressing present actions, but can also be applicable to describing actions that have taken place in the past. For example, “each domain [is] being divided” would express an action taking place at present, whereas “each domain [was] being divided” would refer to an action in the past.
77. I agree that, technically, feature (iii) appears to be related to the generation of the binary grid as suggested by the Opponent. However, on purposive construction, since claim 1 defines the security device and not the method of producing it, I consider that this feature defines how, when the security device was produced, the pixels of the synthetic images were being distributed to the domains that make up the binary grid, so that the synthetic images can be projected through the process of sampling. In other words, at the time “each [synthetic] image is processed digitally” (see feature (i)), the binary grid with its pixels already exists in the security device. I consider that feature (iii) should be interpreted as defining that the pixels of each synthetic image were being distributed to the same location within each digitised domain when the image layer was generated.
78. Feature (iv) further qualifies feature (iii) in that “the pixels in each digitally processed image [were] being distributed … such that each location within one digitized domain is marked with the color of a pixel from a different digitally processed image”. This feature defines in more detail how the synthetic images were being distributed to the digitised domains. I have already discussed that and concluded that the only way in which a location within a digitised domain can be marked with the colour of a pixel from an image is by making the layer image pixel which was assigned that location in the domain to have the same colour as the pixel from the image. Therefore, I consider that this feature should be interpreted as defining that when the image layer was generated and the pixels of each synthetic image were being distributed to the same location within each digitised domain, this was done such that each location within one digitised domain has been assigned to a layer image pixel of the same colour as a pixel from a different synthetic image.
79. In light of the above analysis, it will be useful to summarise my interpretation of claim 1:
1. A security device for projecting a collection of synthetic images, which comprises:
a collection of focusing elements, with each focusing element having an optical footprint; and at least one image layer,
the collection of focusing elements and the at least one image layer together projecting a different synthetic image as the device is viewed at different angles,
wherein, the at least one image layer includes a binary grid of distributed digital images,
the binary grid comprising a plurality of layer image pixels
where each layer image pixel in the binary grid is uniformly coloured in either one of two possible colours,
wherein, the binary grid of the at least one image layer is made up of an array of discrete digitized domains,
each domain constituting an identical or substantially identical subset of each focusing element’s optical footprint,
the domains being discrete in that no two subsets overlap and every point in each subset is closest to its respective focusing element,
each domain being divided into a number of layer image pixels equal to the number of synthetic images with
each layer image pixel assigned a location within the respective domain,
wherein, each synthetic image is created digitally by the process of sampling layer image pixels from the binary grid,
the ratio of the total number of focusing elements divided by the number of pixels in each synthetic image is a constant greater than or equal to one,
the pixels in each synthetic image were being distributed to the same location within each digitized domain when the image layer was generated
such that each location within one digitized domain has been assigned to a layer image pixel of the same colour as a pixel from a different synthetic image,
allowing for the device to project a different synthetic image as the device is viewed at different angles.
I have already explained the meaning I ascribe to the terms “synthetic image” and “layer image”.
80. Despite the somewhat lengthy analysis required to achieve this, I was able to provide unambiguous interpretation of the scope of claim 1. Therefore, I have no reasons to conclude that claim 1 is not clear.
Interpretation of claim 2
81. Claim 2 is directed to “[t]he security device of claim 1” and it further defines that the security device “projects a collection of grayscale or halftone synthetic images, wherein each image is a grayscale image with a reduced color palette” and that “a cluster of focusing elements is used for each image pixel”. It is clear that the first instance of the term “image” refers to the synthetic images and I have no reasons to consider that the remaining instances refer to different images. Therefore, the term “image pixel” should refer to the synthetic image pixels.
82. Mr Lundvall does not disagree with such interpretation:
“It appears to me that the term ‘image pixel’ is intended to refer to a component of the synthetic image which is seen by an observer although there is nothing in claim 1 or claim 2 which defines the term in this way. In this regard I note that claim 6 refers to ‘viewpoint image pixels’ which appears to be consistent with this interpretation.” (Lundvall-1 at [55])
83. Importantly, claim 2 defines that a cluster of focusing elements (which I consider to mean a group of more than one neighbouring focusing elements) is used for (i.e. to project) each synthetic image pixel. In this case the total number of the focusing elements will be larger than the number of pixels in each synthetic image, hence the ratio defined in claim 1 will be greater than one. I already discussed this situation and concluded that, in order to form a pixel of the synthetic image, each lens in the group or cluster must project the same colour. Hence, as in claim 1, each synthetic image pixel will be in one of the two possible colours that the layer image pixels can have. In other words, the arrangement of claim 2 does not provide a mechanism of producing greyscale or halftone synthetic images that is different to the one that could be employed with the arrangement of claim 1, e.g. the greyscale dithering mentioned in the description. Nonetheless, I consider that the scope of claim 2 is clear and unambiguous. Using a cluster of focusing elements to project one synthetic image pixel could be a design option for achieving a specific visual effect and I have no reasons to consider such interpretation of claim 2 unreasonable.
84. In addition, claim 1 does not define the type of the synthetic image that the security device is projecting, i.e. the device of claim 1 is not limited to projecting greyscale images with a reduced colour palette (e.g. the synthetic images of claim 1 can be pure black and white images). Claim 1 also does not limit the synthetic image design to the use of a cluster of focussing elements for projecting each synthetic image pixel. Therefore, the scope of claim 2 is clearly different to the scope of claim 1 and I have no reasons to consider claim 2 redundant.
Interpretation of claim 6
85. With respect to this claim, Mr Lundvall comments that “[c]laim 6 is largely directed to a method for producing the device as claimed in claim 1” (Lundvall-1 at [60]). I note that, in this statement, Mr Lundvall refers to claim 1 as accepted and before the amendment. Mr Merchant is of similar opinion:
“Claim 6 relates to a computer implemented method for producing a security device comprising a number of specified steps. This claim is related to claim 1 and sets out the steps that are undertaken to produce a security device having the features of claim 1. I note that claim 6 refers to ‘digitally processing each of the different raw viewpoint images to form binary images’. I understand by this that claim 6 also incorporates a similar feature to that of claim 1 of the proposed claims in which the image is formed from binary pixels, which are either ‘on’ or ‘off’.” (Merchant at [85], emphasis added)
86. In his second declaration, Mr Lundvall does not appear to disagree.
87. The preamble of claim 6 defines “[a] computer-implemented method for producing a security device”, and according to the Opponent, this means that all steps of the claim must be computer-implemented. On purposive constriction in the context of the claimed invention, I consider that the phrase “computer-implemented method for producing a security device” (i.e. for producing a physical object) only means that a computer is used during the implementation of the method. This however does not mean that the method must, in its entirety, be performed on a computer. As long as a part of the method is performed on a computer, this would satisfy the limitation in question.
88. Claim 6 defines the following (for easier reference, I have labelled the two features of interest):
“choosing a domain for each focusing element in the focusing element sheet, and arranging the domains in the form of a grid on or within the at least one image layer,
the domains constituting identical subsets of each of the focusing element’s optical footprints such that no two subsets overlap and
(a) every point in each subset is closest to its respective focusing element once the at least one image layer is placed in a focal plane of the collection of focusing elements,
(b) wherein exact registration between the domains and the focusing elements is not required”.
89. It is clear that the domains form a grid on or within the image layer. In addition, these domains are each constituting identical subsets of each of the focusing element’s optical footprints. In light of this, feature (a) clearly imposes certain degree of registration between the image layer and the collection of focusing elements, because when the image layer is placed in the focal plane of the collection of focusing elements, every point in each subset (i.e. in each domain of the image layer) is closest to its respective focusing element.
90. While the expression “exact registration” on its own could be vague and imprecise, in light of feature (a), the use of this expression in feature (b) becomes meaningful. Indeed, the exact (e.g. perfect) registration between the domains and the focusing elements is not required as long as the condition specified in feature (a) is satisfied. I note that feature (a) provides a clear boundary of the allowable miss-registration, i.e. starting from a perfect registration the collection of focusing elements can be shifted with respect to the domains of the image layer, but only to the degree that no point in a domain is closer to another (e.g. neighbouring) focusing element than to its corresponding or respective focusing element.
91. With respect to the features equivalent to features of claim 1, following the same reasoning as with respect to claim 1, I will summarise my interpretation of claim 6 as:
6. A computer-implemented method for producing a security device for projecting a collection of synthetic images, wherein the security device is made up of
a collection of focusing elements in the form of a focusing element sheet, each focusing element having an optical footprint; and at least one image layer,
the focusing elements and the at least one image layer together projecting a different viewpoint synthetic image as the device is viewed at different angles,
the method comprising:
(a) forming the at least one image layer by:
(i) compiling a collection of different raw viewpoint images with each raw viewpoint image prescribing what an observer should see when viewing the security device from a given angle;
(ii) choosing a domain for each focusing element in the focusing element sheet, and arranging the domains in the form of a grid on or within the at least one image layer,
the domains constituting identical subsets of each of the focusing element’s optical footprints such that no two subsets overlap and every point in each subset is closest to its respective focusing element once the at least one image layer is placed in a focal plane of the collection of focusing elements,
wherein exact registration between the domains and the focusing elements is not required;
(iii) digitizing each chosen domain by dividing each domain into a number of layer image pixels, which will each represent a portion of a different viewpoint synthetic image,
the number of layer image pixels in each digitized domain being equal to the number of different viewpoint synthetic images,
the digitized domains forming a raster grid;
(iv) digitally processing each of the different raw viewpoint images to form binary synthetic images,
wherein the ratio of the total number of focusing elements in the focusing element sheet that will be used to represent the desired complete synthetic image divided by the number of pixels in each digitally processed viewpoint synthetic image is a constant greater than or equal to one;
(v) distributing the collection of different processed viewpoint synthetic images into the raster grid by making each digitized domain layer image pixel the same color as the color of the corresponding viewpoint synthetic image pixel through a process called ‘distribution’, which involves
assigning an address to each layer image pixel in each digitized domain, and then
assigning one synthetic image to each layer image pixel having the same address in each digitized domain, such that each location within one digitized domain will have a layer image pixel having the same color as the color of a pixel from a different processed viewpoint synthetic image; and
(b) placing the formed image layer or layers in a focal plane of the focusing element sheet.
Interpretation of claim 8
92. My discussion with respect to claim 2 is equally applicable to claim 8, and my interpretation of claim 8 is summarised below:
8. The computer-implemented method of claim 6,
which produces a security device that projects a collection of grayscale or halftone synthetic images, wherein each raw viewpoint image is a raw grayscale or halftone viewpoint image,
wherein the raw viewpoint images are modified by reducing the number of shades of gray in each image’s color palette, optionally dithering the remaining shades of gray in each image’s color palette, and then
representing each such processed viewpoint image as a finished binary synthetic image,
wherein after distributing the collection of different finished binary synthetic images into the raster grid, a cluster of focusing elements is used for each viewpoint synthetic image-pixel.
Clarity summary
93. Based on their interpretation of the claims, the Opponent asserts that claims 2 and 8 (and possibly claim 1) are not clear. In the above discussion, I was able to interpret the claims in question clearly and unambiguously. I am not satisfied that any one of claims 1 to 20 is not clear.
Novelty under Subparagraph 18(1)(b)(i)
94. The well-established test for novelty can be found in Meyers Taylor Pty Ltd v Vicarr Industries Ltd [1977] HCA 19 at [20]; 137 CLR 228 at 235:
“The basic test for anticipation or want of novelty is the same as that for infringement and generally one can properly ask oneself whether the alleged anticipation would, if the patent were valid, constitute an infringement.”
95. This test requires that all essential features of the claimed invention are disclosed in the prior art document. The level of disclosure in the prior art document was considered in The General Tire & Rubber Company v The Firestone Tyre and Rubber Company Limited and Others [1972] RPC 457 at 486 as “… must contain clear and unmistakeable directions to do what the patentee claims to have invented”.
96. The Opponent relies on a single prior art document for novelty and this is discussed below.
Novelty in light of document D2 (WO 2010/094691 A1)
97. Document D2 (also referred to as WO’691 or Rolling Optics) was filed in evidence as exhibit AL-3, and I note that Mr Lundvall is the first named inventor (see also Lundvall-1 at [21]). The document discloses “devices for synthetic integral images and computer-assisted manufacturing thereof” (page 1, lines 4-5). The devices can be used “as security labels on bank-notes or other valuable documents, identification documents etc.” (page 1, lines 10-11). The design and operation of the devices are best explained with reference to the drawings as reproduced below.
98. In understanding the disclosure of document D2, I found that the text in the document is very revealing, hence I will reproduce several passages:
“Fig. 1A illustrates a partial cross section view of an integral image device 10, comprising a polymer foil 5 and giving scenes of an integral image when viewed from one side in different directions. The thickness direction of the polymer foil is denoted by 6. The integral image device comprises a focusing element array 14 of microlenses 12 in a focusing element plane 16. A microlens 12 is an example of a focusing element 11. Other types of focusing elements 11, such as e.g. curved mirrors are also possible to use, as described further below. Below (as defined in the figure) the focusing element plane 16, the integral image device 10 comprises an image plane 26, at which structures 22 that are optically distinguishable are provided. The structures 22 are in the present embodiment embossed structures 21 in an interface 23. However, the structures 22 can in alternative embodiments e.g. comprise printed structures.” (page 4, lines 1-8, emphasis added)
“Each microlens 12 has a respective cell 24 at the image plane 26. The cells 24 are distributed according an image array 21. In the present embodiment, the respective cell 24 is situated straight below the corresponding microlens 12, along the thickness direction 6. Such a configuration is illustrated by Fig. 1B, where a portion of an integral image device 10 is shown from above. In this embodiment, the microlenses 12 are provided in a closed-packed array, forming hexagonal borders 13. In the present embodiment, the cells are of the same size as the microlenses 12 and furthermore aligned therewith in the lateral direction. Borders 25 between the adjacent cells 24 are therefore situated exactly below the borders 13 between the microlenses 12.” (page 4, lines 19-25, emphasis added)
Inventive step summary
142. Following the above analysis, I am not satisfied that the Opponent has established that the invention claimed in either one of claims 1 and 6 is obvious. Claims 2 to 5 and 7 to 20 are ultimately appended to either claim 1 or claim 6 and add further features to those previously defined, hence I have no reason to consider that any one of these claims is obvious. Therefore in conclusion, on the balance of probabilities, I am not satisfied that any one of claims 1 to 20 does not involve an inventive step.
Usefulness (Utility) under Paragraph 18(1)(c)
143. The Opponent asserts that:
“The objectives of AU’640 [the Application] are set out in paragraph [0008] of the specification at [sic] noted in paragraph 51 above. The four objectives are expressed cumulatively. If a patent specification contains a composite promise for an invention, a failure to attain any one of the elements of the composite promise in any claim will render that claim lacking in utility (ESCO Corporation v Ronneby Road Pty Ltd [2018] FCAFC 46). Having regard to the manner in which the specification of AU’640 describes the limitations said to be associated with the prior art and the advantages said to follow from the invention, it is clear that the specification makes a composite promise for that invention including each of the four objectives set out in paragraph [0008].
Claim 1 of AU’640 encompasses security devices which fail to achieve any of the four objectives set out in the specification (Lundvall 1 [54]).” (OS at [103]-[104], emphasis added)
“Even if the specification of AU’640 [the Application] is considered not to set out a composite promise for the invention, at least one promise of the invention is that the finished synthetic image can be designed such that there is no ‘snap’ in the field of view. At Merchant [61] it is said that a major advance of the alleged invention of AU’640 is that it is not necessary to achieve exact registration of the image layer with the lens layer and that the issue of flip does not arise due to the use of mathematical scalar functions. At Merchant [63] it is said that the approach described in AU’640 overcomes the limitations of the prior art which are said to include that the finished synthetic image can be designed such that there is no ‘snap’ in the field of view. At Merchant [71] it is said that the method of eliminating snap described in AU’640 is a significant new improvement.” (OS at [106], emphasis added)
“It is not apparent that any claim of AU’640 [the Application] defines a security device or a method for producing a security device which meets each of the four promises made for the invention. Accordingly the invention as claimed in each of claims 1 to 20 fails to meet the promise of the invention and is lacking in utility.
Alternatively, the only claims of AU’640 which require the use of mathematical scalar functions to produce a synthetic image which has not snap or flip are claims 4 and 10. Accordingly the invention as claimed in each of claims 1 to 3, 5 to 9 and 11 to 20 fails to meet the promise of the invention and is lacking in utility.” (OS at [108]-[109], emphasis added)
The Opponent refers to the evidence of Mr Lundvall as reproduced below:
“As discussed above, the invention of AU’640 [the Application] is said to provide four major improvements over the prior art. I consider that a security device having the features of claim 1 would not necessarily achieve any, let alone all of the alleged improvements. Claim 1 does not require the synthetic images to be 3D, does not require the image to be designed to have no snap, does not require the images to be grey scale or halftone images and does not specify the source of the digital images. I also note that in paragraph [0063] of AU’640 it is said that by eliminating snap ‘the requirement for lens to domain registration is no longer required, and the manufacturability of such devices is enabled using current techniques’. This suggests that the inventors of AU’640 considered that the available manufacturing techniques did not allow the necessary lens to domain registration. If that were so, a useful device having no snap could only be produced if the mathematical scalar functions described in AU’640 were used.” (Lundvall-1 at [54], emphasis added)
“In paragraph [61] of Merchant, it is claimed that the invention disclosed in AU’640 represents and improvement over what is disclosed in the prior art, in particular because the use of the described mathematical scalar functions avoids the phenomenon of ‘flip’ and removes the requirement for exact registration between the image layer and the lens layer. For the reasons discussed in my First Declaration and elaborated below, I do not consider that the invention of AU’640 constitutes such an improvement over the prior art. Further, while the specification of AU’640 does describe the use of such mathematical scalar functions, as discussed in my First Declaration, the majority of the claims of AU’640 are not limited to the use of such functions.” (Lundvall-2 at [16], emphasis added)
144. I will first discuss the issue of possible connection between the absence of a “snap” and the removal of the requirement for (exact) registration between the image layer and the collection of focusing elements. As far as the claimed invention is concerned, the feature “wherein exact registration between the domains and the focusing elements is not required” is defined in claim 6. I have already construed this claim on the basis of the clear and unambiguous meaning of its wording, and I note that my interpretation of the feature in question is in no way connected to the presence or absence of a “snap”. It also appears to me that, in general, a device of the claimed type can exhibit a “snap” even with a perfect registration, any miss-registration affecting only the viewing angle at which the “snap” occurs. Hence, even if the absence of a “snap” might potentially somewhat relax the registration requirements, I can see no reason to conclude that removing the requirement for exact registration as claimed is necessarily directly related to the absence of a “snap” which apparently depends on the specific design of the collection of synthetic images. In addition, on purposive construction of the clear and unambiguous language used in claim 6, it is impermissible for me to include within the scope of the claim any possible limitations that may or may not be inferred by reading the description.
145. The second important issue is whether the available manufacturing techniques did or did not allow the necessary lens to domain registration in order to produce the device claimed e.g. in claims 1 and 6. The body of the specification comments:
“Moreover, in an exemplary embodiment, a special symmetry is used to generate the synthetic images, which allows the device to be manufactured without regard for registration, which is a known problem with prior art devices.” (at [10], emphasis added)
“Device 1 shown in FIG. 24 will result in the range being projected in a direction that is perpendicular to the surface of the device with snap occurring when the device is viewed from high angles. This is the ideal scenario for a device such as the one that was previously described and projects a static 3D object. However, with equal likelihood, current manufacturing processes will result in something like device 2, where the boundary between digitized domains lies directly beneath each lens. This will result in a snap of the 3D image when an observer is looking at the device from a perpendicular location. The range (and copies of the range) are being projected in directions that are only seen at awkward viewing angles. This is undesirable.
In the exemplary embodiment described below, the inventive security device projects images having no snap. In eliminating snap, the requirement for lens to domain registration is no longer required, and the manufacturability of such devices is enabled using current techniques.” (at [62]-[63], emphasis added)
146. Indeed, literal interpretation of some of these statements in the description could potentially lead to an understanding that the available manufacturing techniques did not allow the necessary lens to domain registration to produce any device with a “snap”, because the “snap” could occur when viewing the device from a perpendicular direction which is undesirable. However, this leads to the question why the “no snap” design is not given much more prominence in the description and why embodiments having a “snap” have been described at all (see e.g. the working example described with reference to figures 7-12 at [36]-[41] as well as the section “Grayscale” at [42]-[48] and the section “3D Images” at [49]-[58]).
147. While discussing the background information in the field, Mr Lundvall explains the registration issues in some detail:
“For each of the processes described above, an important factor in the quality of the synthetic image projected to an observer is accurate registration between the image and lens arrays. For the integral image processes it is sometimes possible to ensure accurate registration by not separating the image array from the lens array once the image array has been recorded. However if the process of producing such a device requires separation of the image array from the lens array used to form it, it is then likely to be important to ensure correct registration when the two arrays are re-combined. For spherical lenses there, lack of registration can take a number of forms. For example, there may be misalignment either vertically or horizontally in the plane of the arrays. Instead of the recorded micro images lying centrally beneath a respective lens, they may be aligned toward an edge of the lens or even at the junction between two lenses. With such misalignment, the area of the image array lying directly beneath the lenses may be essentially blank such that when viewed from an angle perpendicular to the array, no object will be visible. In such a situation, the synthetic three-dimensional object will only be visible at more acute viewing angles. Rotational misalignment between the image array and lens array may result in a blurring and also magnification of the synthetic image, similar to a Moire effect.
Registration issues can also arise in the context of lenticular images. Horizontal misalignment in the direction perpendicular to the cylindrical lenses will result in the incorrect image in the image sequence being observed when the device is viewed at an angle perpendicular to its plane. For some image sequences this may not be an issue but for others it may. For example, as discussed above, lenticular images may be affected by flip [apparently the same as ‘snap’]. If the misalignment between the image array and the lens array is half the width of the lenses, the viewing angle at which the ‘flip’ occurs will be perpendicular to the plane of the device which may be particularly undesirable. Of course if the image sequence has been designed to be cyclical as discussed above, such ‘flip’ may not be noticeable, effectively eliminating the need to ensure horizontal registration between the image and lens arrays.” (Lundvall-1 at [18]-[19], emphasis added)
148. Based on the above discussion, on balance, I consider it more likely than not that the available manufacturing techniques did allow, within certain tolerances, the required registration. As I already mentioned, at times the description uses vague and imprecise language which should not be interpreted literally. In addition, Mr Lundvall’s later evidence (see Lundvall-1 at [54] quoted above), does not really suggest that “the available manufacturing techniques did not allow the necessary lens to domain registration” – he was only referring to the wording in the description using the phrase “[i]f that were so” which does not mean the he agrees with the proposition. In conclusion, I am not satisfied that the device claimed in claim 1 or the method claimed in claim 6 cannot be realised using the available manufacturing techniques, even though these claims do not require the absence of a “snap” achieved through the use of the described mathematical scalar functions.
149. Returning to the promise of the invention, I have already quoted paragraph [8] of the description earlier in this decision, however for convenience, I will reproduce the relevant passage of this paragraph below (emphasis added):
“This approach will allow for the following major improvements over the prior art:
(a) The finished synthetic image can be, among other things: a moving 3D object; a dynamic (morphing or transforming) 3D object; a dynamic design of curves, abstract designs, shapes, photographs, 3D objects and images;
(b) The finished synthetic image can be designed such that there is no ‘snap’ in the field of view;
(c) The finished synthetic images can include ‘halftone’ effects similar to grayscale dithering. Furthermore, this method will help enable the coordination of several layers to ultimately produce synthetic images that incorporate full color dynamic designs and 3D images; and
(d) The finished synthetic images do not have to be sourced from models of 3D objects. The synthetic images can originate from any type of digital image (e.g., photographs, drawings, mathematical plots and curves, etc.).”
150. In my view, on fair reading, the above quotation clearly suggests that what is promised is not what the finished synthetic images will be, but instead what they can be. There is no evidence on file to suggest that (e.g.) the finished synthetic images cannot be any one of a moving 3D object, a dynamic (morphing or transforming) 3D object, a dynamic design of curves, abstract designs, shapes, photographs, 3D objects and images. There is also no evidence that the finished synthetic image cannot be designed such that there is no “snap” in the field of view. The same applies to the other advantages. Ultimately, the finished synthetic images will depend on the design of the particular device. The claimed invention enables a device to be designed in a way to achieve any of the above advantages.
151. Based on the evidence on file, on the balance of probabilities, I am not satisfied that the Opponent has established that the invention defined in any one of claims 1 to 20 is not useful.
Clear Enough and Complete Enough Disclosure under Paragraph 40(2)(a)
152. The Explanatory Memorandum to the Intellectual Property Laws Amendment (Raising the Bar) Bill 2011 explains that (emphasis added, reference(s) omitted):
“This item amends paragraph 40(2)(a) by imposing the requirement that a patent specification must disclose the invention in a manner which is clear enough and complete enough for the invention to be performed by a person skilled in the relevant art. This is intended to align the disclosure requirement with that applying in other jurisdictions with the effect that sufficient information must be provided to enable the whole width of the claimed invention to be performed by the skilled person without undue burden, or the need for further invention. This more clearly reflects a fundamental principle of the patent system: in exchange for the exclusive rights given to the patentee, the patentee must share with the public the information necessary to make and use the invention.”
“There are two aspects to this requirement:
·the specification must make the nature of the invention plain; and
·the specification must make it plain how to make or perform the invention.”
“The intention is that paragraph 40(2)(a) be given, as close as is practicable, the same effect as the corresponding provisions of UK legislation and the European Patent Convention.”
153. The test for clear enough and complete enough disclosure is provided in Kirin-Amgen Inc v Hoechst Marion Roussel Limited [2004] UKHL 46; [2005] RPC 9 at [103]:
“Whether the specification is sufficient or not is highly sensitive to the nature of the invention. The first step is to identify the invention and decide what it claims to enable the skilled [person] to do. Then one can ask whether the specification enables [them] to do it.”
154. It was further stated in H. Lundbeck A/S v Generics (UK) Ltd [2008] EWCA Civ 311; [2008] RPC 19 at [29] that:
“In order to decide whether the specification is sufficient, it is … first necessary to decide what the invention is. That must be found by reading and construing the claims, in which the inventor identifies what he claims to be his invention.”
155. The Opponent outlines two main deficiencies. The first one relates to the use of mathematical scalar functions to avoid a “snap”:
“For the reasons set out above, claims 1 to 3, 5 to 9 and 11 to 20 encompass security devices and methods for their manufacture which involve the use of mathematical scalar functions as described in the specification of AU’640 [the Application]. However these claims are not limited to the use of such functions.
At least some of the advantages of the invention disclosed in the specification are said to arise as a result of the use of mathematical scalar functions to eliminate flip or snap (Merchant [61]). As disclosed in the specification of AU’640 at [0063], by eliminating snap ‘the requirement for lens to domain registration is no longer required, and the manufacturability of such devices is enabled using current techniques’. The inventors of AU’640 considered that the available manufacturing techniques did not allow the necessary lens to domain registration to avoid the possibility of avoid flip [sic], absent the use of mathematical scalar functions (Lundvall 1 [54]).
None of claims 1 to 3, 5 to 9 or 11 to 20 requires the use of mathematical scalar functions as described above. The specification provides no disclosure of how a security device which avoids flip can be manufactured other than by the use of mathematical scalar functions. Accordingly, the complete specification of AU’640 does not provide an enabling disclosure of the full scope of subject matter falling within the scope of claims 1 to 3, 5 to 9 or 11 to 20.” (OS at [114]-[116], emphasis added)
156. It is worth noting that the term “snap” is mentioned explicitly only in claims 4, 5, 10, and 11; and from these only claims 4 and 10 are limited to projecting synthetic images with “no snap”. These latter two claims are exactly the claims for which, apparently, the Opponent considers that the complete specification of the Application does provide an enabling disclosure.
157. I have already discussed the possible relationship between the “snap” and the requirement for lens to domain registration and concluded that no such relationship exists within the scope of claim 6. I have also discussed whether the available manufacturing techniques would or would not allow the necessary lens to domain registration, and concluded that the existence of such severe technological limitations cannot be established on the evidence.
158. I consider that the nature of the invention is reflected in the broadest claims 1 and 6 which are directed to a security device and a method of its producing, respectively. I have already construed these claims and, based on my interpretation, I consider that the invention is an optical security device having an image layer and a collection of focusing elements each sampling a small area of the image layer. The device is characterised by the particular way of structuring the image layer and the specific interaction between the image layer and the collection of focussing elements as defined, in order to project a collection of synthetic raster images, the pixels of each synthetic raster image being projected by the focusing elements and being completely independent of the pixels of the other synthetic raster images. Therefore, I consider that the invention claims to enable the person skilled in the art to create an optical security device with particular characteristics as mentioned above and not a device that will achieve any specific type of optical effect. The evidence on file is insufficient for me to conclude that the person skilled in the art reading the specification of the Application will be unable to implement the optical security device of the invention.
159. I note that due to the complete independence of the synthetic images from each other, a vast variety of optical effects is achievable in the claimed security device through appropriate design of the image layer and the corresponding collection of focusing elements. While several types of optical effects are described and claimed in the dependent claims, I do not consider that they are anything more than examples of what can be achieved by the invention. The Opponent has focused their attention on one of these examples, however it is not unreasonable to assume that there also exists an optical effect achievable with the claimed device (hence within the scope of the claimed invention) which is not described in the Application at all. I do not believe that this makes the specification deficient because, as I already mentioned, I do not consider that the invention is directed to any specific optical effect or type of optical effects. While the invention may allow the development of devices exhibiting specific optical effects, the invention itself is not about the effects, but the underlying structure of the device.
160. The second issue is related to the sampling of pixels from the binary grid:
“Feature (ix) of claim 1 requires that each image is processed digitally by sampling pixels from the binary grid. The specification of AU’640 [the Application] provides no disclosure of sampling pixels from the binary grid as part of the step of digitally processing images. Rather, as described in AU’640, sampling pixels from the binary grid is part of the process of viewing a synthetic image projected by the device. There is nothing in the specification which would enable a skilled addressee to digitally process a raw viewpoint image by sampling pixels from the binary grid. Accordingly, the complete specification of AU’640 does not provide an enabling disclosure of the full scope of subject matter falling within the scope of claim 1 or of any of claim when dependent on claim 1.” (OS at [117], emphasis added)
161. I consider that the above issue is clearly related to the Opponent’s interpretation of this feature of claim 1. In my interpretation of claim 1, I concluded that the feature in question means that each synthetic image is created digitally by the process of sampling layer image pixels from the binary grid. I also commented that since the synthetic image is created digitally by the process of sampling, it is then a digitally processed image; and I was able to interpret the related features in claim 1 on this basis. I am not satisfied that, on proper construction, this second issue actually exists.
162. Following the above discussion, on the balance of probabilities, I am not satisfied that the Opponent has established that clear enough and complete enough disclosure is lacking.
Support by Matter Disclosed in the Specification under Subsection 40(3)
163. The Explanatory Memorandum to the Intellectual Property Laws Amendment (Raising the Bar) Bill 2011 explains that (emphasis added, reference(s) omitted):
“Broadly speaking, the terms ‘support’ and ‘full support’ pick up two concepts:
·there must be a basis in the description for each claim; and
·the scope of the claims must not be broader than is justified by the extent of the description, drawings and contribution to the art.”
“This item is intended to align the Australian requirement with overseas jurisdictions’ requirements (such as the UK). Overseas case law and administrative decisions in respect of the ‘support’ requirement will be available to Australian courts and administrative decision-makers to assist in interpreting the new provision.”
164. The Opponent again outlines two issues, corresponding to the issues raised with respect to the ground of clear enough and complete enough disclosure. With respect to the first issue, after discussing the evidence of Mr Merchant in light of the disclosures of documents D2 and D3, the Opponent concludes that ‘[t]o the extent that AU’640 [the Application] makes a technical contribution to the art, it is by the use of continuous mathematical scalar functions to avoid snap” (OS at [123]). The Opponent continues:
“The technical contribution to the art as disclosed in the specification of AU’640 [the Application] is the ability to form such devices using available manufacturing techniques while avoiding flip by the use of mathematical scalar functions. The beneficial effect said to be achieved by the invention is derived from the specific use of such functions (Merchant [61]).
None of None of [sic] claims 1 to 3, 5 to 9 or 11 to 20 require the use of mathematical scalar functions as described above. The specification provides no disclosure of how a security device which avoids flip can be manufactured other than by the use of mathematical scalar functions. Accordingly, none of claims 1 to 3, 5 to 9 or 11 to 20 are supported by the technical contribution to the art.” (OS at [126]-[127], emphasis added)
165. It appears that the Opponent places a lot of emphasis on the evidence of Mr Merchant to establish the contribution of the invention. However, I note that Mr Merchant also states that:
“At paragraph [0011] it is noted that in contrast to a moire magnifier, which has an array of more or less ‘continuous’ images, the image layer in the security device of the Opposed Application is a binary grid of distributed digital images where each pixel in the grid is either ‘on or off’ (i.e coloured or uncoloured). This can be seen, for example, in Fig 8, frame 2 where each numeral is formed as a combination of black and white square pixels. This is a particular feature of this document as compared to all the other documents I have considered which all provide ‘continuous’ digital images rather than images formed from discrete pixels.” (Merchant at [66], emphasis added)
“This method of achieving grayscale is not discussed in any of the previous documents I have reviewed.” (Merchant at [74]).
It appears to me that, although Mr Merchant indeed emphasises the ability to avoid a “snap”, he does not consider that this is the only distinction between the present invention and the prior art.
166. In addition, I have not established that claims 1 and 6 are not novel or not inventive. Therefore, a contribution to the art is made by the invention as claimed in these claims. As I have already discussed, neither of these claims is related to the “snap” or its avoidance. I have also noted that the evidence on file is insufficient for me to conclude that the person skilled in the art reading the specification of the Application will be unable to implement the optical security device of the invention as claimed in claims 1 and 6. The entire width of these claims can be traversed by designing different optical effects via selecting appropriate collections of synthetic images, some of which are described in the specification and defined in the dependent claims.
167. Regarding the second issue, the Opponent explains:
“Feature (ix) of claim 1 requires that each image is processed digitally by sampling pixels from the binary grid. The specification of AU’640 [the Application] provides no disclosure of sampling pixels from the binary grid as part of the step of digitally processing images. Rather, as described in AU’640, sampling pixels from the binary grid is part of the process of viewing a synthetic image projected by the device. The technical contribution provided in the specification is non-existent with respect to how to digitally process a raw viewpoint image by sampling pixels from the binary grid. Accordingly, claim 1 and any claim which is dependent on claim 1 is not supported by the technical contribution to the art.” (OS at [128], emphasis added)
168. As in the case of the equivalent issues raised for the ground of clear enough and complete enough disclosure, I consider that the above issue is clearly related to the Opponent’s interpretation of this feature of claim 1 which is different from my interpretation. I am not satisfied that, on proper construction, this second issue actually exists.
169. Following the above discussion, on the balance of probabilities, I am not satisfied that the Opponent has established that any one of the claims is not supported by matter disclosed in the specification.
Conclusion and Costs
170. I have decided that, for the purposes of deciding the opposition, I will not consult under the provisions of regulation 5.23 the New Evidence, i.e. the declaration by Mr Gregory John Whitehead dated 27 August 2018 with exhibits GJW-1 to GJW-4.
171. I have not established that any one of the claims is not novel, does not involve an inventive step, is not clear, or is not supported by matter disclosed in the specification. I have also not established that the invention defined in any one of the claims is not useful or that the specification does not disclose the invention in a manner which is clear enough and complete enough for the invention to be performed by a person skilled in the relevant art. It follows that the opposition is unsuccessful on all grounds.
172. With respect to the award of costs, the Opponent submits:
“The Opponent seeks its costs in this matter.
As noted above, the specification of AU’640 [the Application] was amended following the filing of evidence in support, including amendments to claim 1. Accordingly, the Opponent submits that in the event that no ground of opposition is upheld, there should be no award of costs, or alternatively, costs should be awarded against the applicant up until the date of the amendment.” (OS at [129]-[130])
173. In contrast, the Applicant submits that:
“Costs should follow the event. Further, we do not agree that, if the opposition is dismissed the Applicant is only entitled to costs after the date of the amendment. There are several grounds of opposition (all of the grounds other than novelty and obviousness) which are not in any way affected by the amendment to claim 1.” (AS at [168])
174. Indeed, it is a normal practice that costs should follow the event and, in the present case, this would mean that the costs should be awarded against the Opponent. However, I note that the Application was amended during the opposition period. While the Applicant asserts that several ground of opposition are not in any way affected by the amendment to claim 1, as I already commented, the language of the specification including the claims is at times vague and imprecise, and I do not consider that the Opponent’s concerns were entirely unreasonable. In the present circumstances, I consider it fair to award costs according to Schedule 8 up to the date of filing the request to amend the specification of the Application against the Applicant, and to award costs according to Schedule 8 from the date of filing the request to amend the specification of the Application against the Opponent.
Dr V. Z. Kolev
Delegate of the Commissioner of Patents
0
3
0