DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Objections
Claim 10 is objected to because of the following informalities: the claim limitation “generating a set of discrete cosine transform (DCT) coefficients for the groups the groups of pixels within each block size” should recite “generating a set of discrete cosine transform (DCT) coefficients for the groups of pixels within each block size” for proper grammar. Appropriate correction is required.
Response to Arguments
Applicant’s arguments, see remarks, filed 01/15/2026, with respect to the rejection of independent claims 1, 8 and 14 under 35 U.S.C. 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground of rejection is made in view of U.S. Patent Publication No.: 10,839,564 (Gueguen et al.) (hereinafter Geuguen). Although Applicant has presented amendments to the above independent claims, the arguments that obviate the previous rejection combination of the 103 rejection of primary reference Viktorovich and secondary reference Shi, were with respect to language in the previous claim set; Upon review of secondary reference, Shi, Examiner agrees with applicant that Shi fails to teach generating a set of features for each block size of a plurality of block sizes. Therefore, this rejection is a second non-final rejection, rather than a final rejection, and Gueguen is introduced to teach the limitation. It appears the amended limitations were introduced to address the 35 U.S.C. 101 rejection and not the 103 rejection.
Applicant's arguments filed 01/15/2026, regarding the rejection of the claims under 35 U.S.C. 101 have been fully considered but they are not persuasive. Applicant argues, on pages 7-8 of the remarks, that “
PNG
media_image1.png
204
786
media_image1.png
Greyscale
PNG
media_image2.png
96
770
media_image2.png
Greyscale
”.
Examiner disagrees. A human easily has the ability to obtain image data with plurality of pixels by simply taking an image with a smartphone or other digital camera; A pixel (short for "picture element") is the smallest, fundamental unit of a digital image or display, appearing as a tiny dot or square of color. Millions of these pixels, typically composed of red, green, and blue sub-pixels, are arranged in a grid to form the images on screens, such as smartphones, televisions, and cameras; therefore, when a human uses a digital camera to take a picture of something with a digital camera (generic machine), then the image data implicitly comprises pixels; further, preprocessing the pixels of the image data to generate preprocessed image data is extremely broad and fails to delineate any specifics of how the pixels are processed that could not be done via the human mind; a human uses their human vision observing generic image editing software on a computer or the same smartphone used to take the digital image to do any sort of “preprocessing”, such as adding filters, cropping, object removal, selective adjustments in color or lighting, or layering, for example; Applicants arguments are only true if there are technical details in the claim as to the preprocessing that a human could not execute themselves using generic computer machines. Further, as discussed in the previous non-final rejection, the human has the ability to split the digital image into different sections/blocks (each section implicitly has a certain number of pixels) with different sizes and then identifying objects (features), for example, in each different size block section of the digital image; as stated above with regards to the preprocessed data, unless some technical aspects that deal with the pixel level in a way that is impossible for human vision to handle, then the claim is still rejectable under 35 U.S.C. 101; all digital images have pixels so dividing an image into blocks does that for the pixels, which is easily done by a human using their own thinking, vision, and generic machines with no technical specificity.
Therefore, the rejection under 35 U.S.C. 101 is maintained.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1, 8, and 14 are rejected are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., an abstract idea) without integration into a practical application or recitation of significantly more.
In the analysis below, the method of independent claim 1 is considered representative of independent claim 14 since all of the independent claims recite identical steps despite being directed to different statutory matter. Furthermore, independent claims 1, 8, and 14 are directed to one of the four statutory categories of eligible subject matter (apparatuses for independent claims 1 and 8 and process for independent claim 14); thus, the claims pass Step 1 of the Subject Matter Eligibility Test (See flowchart in MPEP 2106).
Step 2A, prong 1 analysis:
Independent claims 1 and 14 are directed to obtaining image data comprising a plurality of pixels; preprocessing the pixels of the image data to generate preprocessed image data; generating, using the preprocessed image data, a set of features for a group of pixels within each block size of a plurality of block sizes; processing the generated sets of features to determine whether the image data is authentic; and when it is determined that the image data is authentic, providing a positive verification indication for the obtained image data.
Each of the above steps can be performed mentally. In particular, a human takes images from a smartphone (digital smartphone images have pixels) and puts a filter (pre-processing) over the images; the human prints multiple copies of the images out and divides the images into 2 sections, 4 sections, 8 sections, etc. up to 16 sections for 4 copies of the images; each divided image is analyzed using human vision, analyzing the features of the images (objects in image, colors, body distortions, strange looking aspects, etc.) to see if it is authentic or not and indicating on the image If the human thinks they are fake or not; pixels and different groups of pixels are implicitly included in all the analysis steps including the preprocessing and generating features steps; therefore, this process can all be done mentally.
Independent claim 8 is directed to preprocessing pixels of the image data to extract luminance data from the image data; generating, using the luminance data, a set of features for a group of pixels within each block size of a plurality of block sizes; processing the generated sets of features to determine whether the image data is authentic; and when it is determined that the image data is authentic, providing a positive verification indication for the image data. Claim 8 is analogous to claims 1 and 14 except for extracting luminance data from the images and using the luminance data to generate a set of features. A human taking images can extract luminance data by indicating how bright or dark an image is, and the brightness/darkness estimated by the human using their own vision is used to determine features in the image such as objects or body parts of people in the image. The remaining limitations are similar to the discussion of independent claims 1 and 14 above; therefore, this process can all be done mentally.
As such, the description in independent claims 1 and 20 is an abstract idea – namely, a mental process. Accordingly, the analysis under prong one of step 2A of the Subject Matter Eligibility Test does not result in a conclusion of eligibility (See flowchart in MPEP 2106).
Additional elements:
The additional element recited in independent claims 1, 8, and 14 are a processor and a memory.
Step 2A, prong 2 analysis:
The above-identified additional elements do not integrate the judicial exception into a practical application.
Each of the other additional elements (processor and memory) amounts to merely using different devices as tools to perform the claimed mental process. Implementing an abstract idea on a computer or using known generic devices does not integrate a judicial exception into a practical application (See MPEP 2106.05(f)).
Moreover, the additional elements of the claims do not recite an improvement in the functioning of a computer or other technology or technical field, the claimed steps are not performed using a particular machine, the claimed steps do not effect a transformation, and the claims do not apply the judicial exception in any meaningful way beyond generically linking the use of the judicial exception to a particular technological environment (See MPEP 2106.04(d)). Therefore, the analysis under prong two of step 2A of the Subject Matter Eligibility Test does not result in a conclusion of eligibility (See flowchart in MPEP 2106).
Step 2B:
Finally, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Each of the other additional elements (processor and memory) are generic computer features which perform generic computer functions that are well-understood, routine, and conventional and do not amount to more than implementing the abstract idea with a computerized system. Thus, taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception (the abstract idea).
Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation, and mere implementation on a generic computer does not add significantly more to the claims. Accordingly, the analysis under step 2B of the Subject Matter Eligibility Test does not result in a conclusion of eligibility (See flowchart in MPEP 2106).
For all of the foregoing reasons, independent claims 1, 8, and 14 do not recite eligible subject matter under 35 USC 101.
Claims 4, 11, and 17 recite wherein processing the generated sets of features is performed using at least one of: a set of rules; a set of thresholds; or a machine learning model trained according to a first set of image data annotated as authentic and a second set of image data annotated as altered. A human uses a set of rules in terms of determining features in an image such as only verifying a feature if the brightness of the feature or object in the image is above a certain threshold of darkness that makes it observable, for example; therefore, this process can all be done mentally.
Claims 5 and 18 recite wherein: the image data is obtained as a request that is received via an application programming interface (API); and the positive verification indication is provided, via the API, in response to the request. An API is generic computing that a human uses to decide to want to obtain images and sends requests through for obtaining; once the evaluation of the authenticity of the images is done by the human using their own vision, then the verification of authentic or not authentic is sent back via the API; the API simply helps facilitate the mental process of the human; therefore, this process can all be done mentally.
Claims 6, 12, and 19 recite wherein the set of operations further comprises: when it is determined that the image data is not authentic, performing additional processing of the image data. If the human decides the image is not authentic then they can spend more time analyzing what parts of the image they believe are the reasoning for the inauthenticity decision and not those aspects in the image itself, as an example of additional processing; therefore, this process can all be done mentally.
Claims 7, 13, and 20 recite wherein the image data is obtained from a remote computing device. A human sends an email request to another database to receive images; the remote computing device is a generic machine facilitating a human mental process of obtaining images from another location or database; therefore, this process can all be done mentally.
Therefore, dependent claims 4-7, 11-13, and 17-20 recite the same abstract idea of a mental process which can be performed in the mind with the aid of pen and paper, and are therefore also rejected under 35 U.S.C. 101.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2, 4, 6, 8-9, 11-12, 14-15, 17, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application Publication No.: 2023/0386003 (Viktorovich et al.) (hereinafter Viktorovich), in view of U.S. Patent Publication No.: 10,839,564 (Gueguen et al.).
Regarding claim 1, Viktorovich teaches a system comprising: at least one processor; and memory storing instructions that, when executed by the at least one processor, causes the system to perform a set of operations, the set of operations comprising: (Viktorovich, para. [0073]:” System 900 may comprise main memory 915. Main memory 915 provides storage of instructions and data for programs executing on processor 910, such as one or more of the functions and/or modules discussed herein.”;
PNG
media_image3.png
488
544
media_image3.png
Greyscale
)
obtaining image data comprising a plurality of pixels (Viktorovich, para. [0029]; FIG. 1; para. [0061]-[0062]; FIG. 8: “An illustration of a source JPEG-compressed image with a quality index of 90 is shown in FIG. 1”;
PNG
media_image4.png
670
431
media_image4.png
Greyscale
;
“FIG. 8 illustrates an example process 800 for detecting and localizing a falsified area in a JPEG image, according to an embodiment. Process 800 may be implemented by a system that is responsible for verifying documents, such as identity documents, in images, which may be acquired by mobile devices (e.g., smartphones, tablet computers, etc.) or uploaded. Initially, in subprocess 810, a JPEG image is received. The JPEG image may be an RGB image that has been subjected to JPEG compression (e.g., lossless JPEG compression). The JPEG image may be an image of an identity document (e.g., passport, drivers license, etc.). There is a possibility that falsifications have been introduced to the JPEG image. For example, a bad actor may use software to replace a portion of the JPEG image (e.g., a text field, image field, etc.) with a copy and pasted portion to thereby create a falsified area in the JPEG image.”;
PNG
media_image5.png
976
602
media_image5.png
Greyscale
);
preprocessing the pixels of the image data to generate preprocessed image data (Viktorovich, para. [0030]; para. [0006]: “1. Complementing the image sides up to the multiplicity with 8 black pixels. 2. Color space transformation from RGB to YCbCr.”; “complementing sides of the JPEG image up to a multiplicity with eight black pixels; transforming the JPEG image from a red-green-blue (RGB) color space to a luma-blue-difference-chroma-red-difference-chroma (YCbCr) color space”);
generating, using the preprocessed image data, a set of features for a group of pixels within a block size of a plurality of blocks (Viktorovich, para. [0030]; para. [0063]: “For the subsequent steps of the algorithm, it is necessary to calculate the DCT coefficients of the image with the preservation of information about the position of the blocks of 8×8 pixels corresponding to the coefficients. This is achieved by the following steps … 3. Breaking the brightness channel in the YCbCr (Y) color space into blocks of 8×8 pixels. 4. Application of discrete cosine transform for each block of 8×8 pixels. 5. Rejecting from consideration blocks in which there are pixels with saturated values (0 or 255) in any RGB channel and blocks in which the brightness (Y) in YCbCr at the moment of transition from RBG to YCbCr went beyond the range [0, 1]. Such pixels will be called saturated. Step 5 is necessary, since the DCT coefficients in such blocks can behave unpredictably and do not fall into the range [kq−1, kq+1], where k∈Z, and q is the quantization step.”; “In subprocess 820, the DCT coefficients of the JPEG image are computed. The DCT coefficients may be computed as described above.”; see step 820 in FIG. 8 above);
processing the generated sets of features to determine whether the image data is authentic; and when it is determined that the image data is authentic, providing a positive verification indication for the obtained image data (Viktorovich, para. [0064]-[0069]: “In subprocess 830, the quantization matrix of the JPEG image is estimated based on the DCT coefficients, computed in subprocess 820. The quantization matrix may be estimated as described above. In subprocess 840, a search is performed for discrepancies between the DCT coefficients, computed in subprocess 820, and the quantization matrix, estimated in subprocess 830. The search may be performed as described above. In subprocess 850, it is determined whether or not the JPEG image contains any falsified areas based on the result of the search performed in subprocess 840. If the JPEG image is not determined to contain any falsified areas (i.e., “No” in subprocess 850), process 800 proceeds to subprocess 860. Otherwise, if the JPEG image is determined to contain a falsified area (i.e., “Yes” in subprocess 850), process 800 proceeds to subprocess 870. In subprocess 860, the JPEG image may be determined to not be falsified. In this case, the output of process 800 may be an indication that the JPEG image is an authentic image of an identity document. It should be understood that process 800 could be one of multiple tests that are used to determine the authenticity of the JPEG image. Alternatively, process 800 could be the only test that is used to determine the authenticity of the JPEG image. In subprocess 870, the JPEG image may be determined to be falsified. In this case, the output of process 800 may be an indication that the JPEG image is an authentic image of an identity document. In addition, in subprocess 880, the falsified area(s) of the JPEG image may be localized as described above. In this case, the output of process 800 may comprise an identification of the location of each area that has been falsified and/or an identification of each field in the identity document that has been falsified.”; see steps 830, 840, 850, 860, 870, and 880 in FIG. 8 above).
Viktorovich fails to teach
generating, using the preprocessed image data, a set of features for each block size of a plurality of block sizes.
Gueguen teaches
generating, using the preprocessed image data, a set of features for each block size of a plurality of block sizes (Gueguen, col. 4, lines 30-64; col. 5, lines 11-16; FIG. 2: “FIG. 2 illustrates an encoding process, in accordance with an embodiment. In particular, the example of FIG. 2 demonstrates a process for compressing an image using JPEG encoding. Pixels of a color image are represented digitally by red, green, and blue (RGB) components 210. Thus, the digital image data may be thought of as a block with height and width of the original image (e.g., measured in pixels), and with a depth of three, wherein each of the three layers includes values for one of the red, green, and blue image components. The RGB 210 representation of the image is converted into a YCbCr representation 220. YCbCr is a color space that is sometimes used instead of the RGB color space to represent color image data digitally. The YCbCr representation of the image data represents the image as a luma (i.e., brightness) component (Y) and two chromatic components (Cb, Cr). The luma component represents brightness within an image, and the chromatic components represent colors. Since human eyesight is more sensitive to differences in brightness than differences in colors, the chroma components may be subsampled to a lower resolution than the luma component, reducing the size of the image data. For example, in FIG. 2, the YCbCr representation 220 of the image data includes a full resolution brightness component and two chroma components with halved height and width dimensions with respect to the original image size. The three channels (luma, chroma, and chroma) are projected through a discrete cosine transform (DCT) and quantized … Given that the DCT coefficients 230 from the luma channel form a differently sized block of data than those of the chroma channels, transformation functions may be applied to some or all of the blocks of DCT coefficients 230 to match the spatial dimensions of the blocks of DCT coefficients 230.”;
PNG
media_image6.png
347
716
media_image6.png
Greyscale
).
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the set of operations, as taught by Viktorovich, to include generating, using the preprocessed image data, a set of features for each block size of a plurality of block sizes, as taught by Gueguen.
The suggestion/motivation for doing so would have been that using multiple block sizes for Discrete Cosine Transform (DCT) coefficients in image processing allows for adaptive compression, optimizing efficiency by using large blocks (e.g., 16x16, 32x32) for smooth, low-detail areas and smaller blocks (e.g., 4x4, 8x8) for high-detail, edge-heavy areas; this approach significantly reduces blocking artifacts, improves image quality at lower bit rates, and enhances energy compaction.
Therefore, it would have been obvious to combine Viktorovich, with Gueguen, to obtain the invention as specified in claim 1.
Regarding claim 2, Viktorovich, in view of Gueguen, teaches the system of claim 1, wherein preprocessing the image data comprises performing at least one of: converting a color space of the image data to YCbCr; or extracting the Y channel of the image data ((Viktorovich, para. [0030]; para. [0063]; see rejection of claim 1 above; Viktorovich teaches converting a color space of the image data to YCbCr).
Regarding claim 4, Viktorovich, in view of Gueguen, teaches the system of claim 1, wherein processing the generated sets of features is performed using at least one of: a set of rules; a set of thresholds; or a machine learning model trained according to a first set of image data annotated as authentic and a second set of image data annotated as altered (Viktorovich, para. [0009]: “Searching for discrepancies between the computed DCT coefficients and the estimated quantization matrix may comprise: generating a first image I1 in which each of the plurality of blocks is assigned a pixel, wherein each pixel for each of the plurality of blocks has a value representing a number of frequencies at which a DCT coefficient corresponding to the block does belong to a range {[k{circumflex over (q)}−1], k{circumflex over (q)}+∈Z}, wherein k is an integer value within a range Z, and {circumflex over (q)} is the estimated quantization step for a respective frequency; determining a set M2 of frequencies, for which at least one of the plurality of blocks at one of the frequencies in set M2 has a computed DCT coefficient not belonging to the range {[k{circumflex over (q)}−1], k{circumflex over (q)}+∈Z}; generating a second image I2 in which each of the plurality of blocks is assigned a pixel, wherein a size of the second image I2 is a same size as the first image I1, and wherein each pixel for each of the plurality of blocks in the second image I2 has a value representing a number of frequencies at which the DCT coefficient corresponding to the block has an absolute value greater than one; and generating a third image I3, with a same size as the first image I1 and the second image I2, according to:
PNG
media_image7.png
50
188
media_image7.png
Greyscale
wherein u is a pixel position, and t is a predefined value”; Viktorovich teaches a set of rules to compare DCT coefficients to a quantization matrix to find differences which indicate tampering/lack of authenticity of the image).
Regarding claim 6, Viktorovich, in view of Gueguen, teaches the system of claim 1, wherein the set of operations further comprises: when it is determined that the image data is not authentic, performing additional processing of the image data (Viktorovich, para. [0068]-[0069]; FIG. 1; para. [0058]-[0060]: “In subprocess 870, the JPEG image may be determined to be falsified. In this case, the output of process 800 may be an indication that the JPEG image is an authentic image of an identity document. In addition, in subprocess 880, the falsified area(s) of the JPEG image may be localized as described above. In this case, the output of process 800 may comprise an identification of the location of each area that has been falsified and/or an identification of each field in the identity document that has been falsified.”; see steps 870 and 880 in FIG. 8 in the rejection of claim 1 above; “Localization of Falsified Areas; An auxiliary image I4 is formed from the image I3, on which morphological disconnection and blurring are additionally applied to eliminate parasitic elements resulting from random deviations of the DCT coefficients. Based on image I4, the maximum is calculated—an estimate of manipulations probability. Image I4 is used to construct image I5 with the sizes of the original image. The pixel value within each 8×8 block of image I5 is equal to the pixel value of image I4 corresponding to this block. Estimation of manipulations probability and image I5 are the result of the algorithm operation. For this example (FIG. 2), the estimate of manipulations probability is 0.68302. Contrasted images I4 and I5 are shown in FIGS. 6 and 7, respectively. thus, the above method for detecting and localizing a falsified area in JPEG images makes it possible to detect some cases of falsification of data in an image without reference to its content and has the potential to be used to verify the authenticity of images of documents of various types.”).
Regarding claim 8, Viktorovich teaches a method for processing image data to determine whether the image data is authentic, the method comprising: (Viktorovich, abstract: “This disclosure relates to verifying the authenticity of images of documents. Embodiments increase the accuracy of determining the falsified area in the image. The method for detecting and localizing a falsified area in JPEG images is to consider, in the RGB color space, an image that was subjected to JPEG compression.)
preprocessing the pixels of image data to extract luminance data from the image data (Viktorovich, para. [0030]; para. [0006]: “1. Complementing the image sides up to the multiplicity with 8 black pixels. 2. Color space transformation from RGB to YCbCr.”; “complementing sides of the JPEG image up to a multiplicity with eight black pixels; transforming the JPEG image from a red-green-blue (RGB) color space to a luma-blue-difference-chroma-red-difference-chroma (YCbCr) color space”);
generating, using the luminance data, a set of features for a group pf pixels within a block size of a plurality of blocks (Viktorovich, para. [0030]; para. [0063]: “For the subsequent steps of the algorithm, it is necessary to calculate the DCT coefficients of the image with the preservation of information about the position of the blocks of 8×8 pixels corresponding to the coefficients. This is achieved by the following steps … 3. Breaking the brightness channel in the YCbCr (Y) color space into blocks of 8×8 pixels. 4. Application of discrete cosine transform for each block of 8×8 pixels. 5. Rejecting from consideration blocks in which there are pixels with saturated values (0 or 255) in any RGB channel and blocks in which the brightness (Y) in YCbCr at the moment of transition from RBG to YCbCr went beyond the range [0, 1]. Such pixels will be called saturated. Step 5 is necessary, since the DCT coefficients in such blocks can behave unpredictably and do not fall into the range [kq−1, kq+1], where k∈Z, and q is the quantization step.”; “In subprocess 820, the DCT coefficients of the JPEG image are computed. The DCT coefficients may be computed as described above.”; step 820 in FIG. 8 below;
PNG
media_image8.png
723
447
media_image8.png
Greyscale
);
processing the generated sets of features to determine whether the image data is authentic; and when it is determined that the image data is authentic, providing a positive verification indication for the image data (Viktorovich, para. [0064]-[0069]: “In subprocess 830, the quantization matrix of the JPEG image is estimated based on the DCT coefficients, computed in subprocess 820. The quantization matrix may be estimated as described above. In subprocess 840, a search is performed for discrepancies between the DCT coefficients, computed in subprocess 820, and the quantization matrix, estimated in subprocess 830. The search may be performed as described above. In subprocess 850, it is determined whether or not the JPEG image contains any falsified areas based on the result of the search performed in subprocess 840. If the JPEG image is not determined to contain any falsified areas (i.e., “No” in subprocess 850), process 800 proceeds to subprocess 860. Otherwise, if the JPEG image is determined to contain a falsified area (i.e., “Yes” in subprocess 850), process 800 proceeds to subprocess 870. In subprocess 860, the JPEG image may be determined to not be falsified. In this case, the output of process 800 may be an indication that the JPEG image is an authentic image of an identity document. It should be understood that process 800 could be one of multiple tests that are used to determine the authenticity of the JPEG image. Alternatively, process 800 could be the only test that is used to determine the authenticity of the JPEG image. In subprocess 870, the JPEG image may be determined to be falsified. In this case, the output of process 800 may be an indication that the JPEG image is an authentic image of an identity document. In addition, in subprocess 880, the falsified area(s) of the JPEG image may be localized as described above. In this case, the output of process 800 may comprise an identification of the location of each area that has been falsified and/or an identification of each field in the identity document that has been falsified.”; see steps 830, 840, 850, 860, 870, and 880 in FIG. 8 above).
Viktorovich fails to teach
generating, using the luminance data, a set of features for a group of pixels within each block size of a plurality of block sizes.
Gueguen teaches
generating, using the luminance data, a set of features for a group of pixels within each block size of a plurality of block sizes (Gueguen, col. 4, lines 30-64; col. 5, lines 11-16; FIG. 2: “FIG. 2 illustrates an encoding process, in accordance with an embodiment. In particular, the example of FIG. 2 demonstrates a process for compressing an image using JPEG encoding. Pixels of a color image are represented digitally by red, green, and blue (RGB) components 210. Thus, the digital image data may be thought of as a block with height and width of the original image (e.g., measured in pixels), and with a depth of three, wherein each of the three layers includes values for one of the red, green, and blue image components. The RGB 210 representation of the image is converted into a YCbCr representation 220. YCbCr is a color space that is sometimes used instead of the RGB color space to represent color image data digitally. The YCbCr representation of the image data represents the image as a luma (i.e., brightness) component (Y) and two chromatic components (Cb, Cr). The luma component represents brightness within an image, and the chromatic components represent colors. Since human eyesight is more sensitive to differences in brightness than differences in colors, the chroma components may be subsampled to a lower resolution than the luma component, reducing the size of the image data. For example, in FIG. 2, the YCbCr representation 220 of the image data includes a full resolution brightness component and two chroma components with halved height and width dimensions with respect to the original image size. The three channels (luma, chroma, and chroma) are projected through a discrete cosine transform (DCT) and quantized … Given that the DCT coefficients 230 from the luma channel form a differently sized block of data than those of the chroma channels, transformation functions may be applied to some or all of the blocks of DCT coefficients 230 to match the spatial dimensions of the blocks of DCT coefficients 230.”;
PNG
media_image6.png
347
716
media_image6.png
Greyscale
).
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the method, as taught by Viktorovich, to include generating, using the luminance data, a set of features for a group of pixels within each block size of a plurality of block sizes, as taught by Gueguen.
The suggestion/motivation for doing so would have been that using multiple block sizes for Discrete Cosine Transform (DCT) coefficients in image processing allows for adaptive compression, optimizing efficiency by using large blocks (e.g., 16x16, 32x32) for smooth, low-detail areas and smaller blocks (e.g., 4x4, 8x8) for high-detail, edge-heavy areas; this approach significantly reduces blocking artifacts, improves image quality at lower bit rates, and enhances energy compaction.
Therefore, it would have been obvious to combine Viktorovich, with Gueguen, to obtain the invention as specified in claim 8.
Regarding claim 9, Viktorovich, in view of Gueguen, teaches the method of claim 8, wherein preprocessing the image data comprises: converting a color space of the image data to YCbCr; and extracting the Y channel of the image data as the luminance data (Viktorovich, para. [0030]; para. [0063]; see rejection of claim 1 above; Viktorovich teaches converting a color space of the image data to YCbCr).
Regarding claim 11, Viktorovich, in view of Gueguen, teaches the method of claim 8, wherein processing the generated sets of features is performed using at least one of: a set of rules; a set of thresholds; or a machine learning model trained according to a first set of image data annotated as authentic and a second set of image data annotated as altered (Viktorovich, para. [0009]: “Searching for discrepancies between the computed DCT coefficients and the estimated quantization matrix may comprise: generating a first image I1 in which each of the plurality of blocks is assigned a pixel, wherein each pixel for each of the plurality of blocks has a value representing a number of frequencies at which a DCT coefficient corresponding to the block does belong to a range {[k{circumflex over (q)}−1], k{circumflex over (q)}+∈Z}, wherein k is an integer value within a range Z, and {circumflex over (q)} is the estimated quantization step for a respective frequency; determining a set M2 of frequencies, for which at least one of the plurality of blocks at one of the frequencies in set M2 has a computed DCT coefficient not belonging to the range {[k{circumflex over (q)}−1], k{circumflex over (q)}+∈Z}; generating a second image I2 in which each of the plurality of blocks is assigned a pixel, wherein a size of the second image I2 is a same size as the first image I1, and wherein each pixel for each of the plurality of blocks in the second image I2 has a value representing a number of frequencies at which the DCT coefficient corresponding to the block has an absolute value greater than one; and generating a third image I3, with a same size as the first image I1 and the second image I2, according to:
PNG
media_image7.png
50
188
media_image7.png
Greyscale
wherein u is a pixel position, and t is a predefined value”; Viktorovich teaches a set of rules to compare DCT coefficients to a quantization matrix to find differences which indicate tampering/lack of authenticity of the image).
Regarding claim 12, Viktorovich, in view of Gueguen, teaches the method of claim 8, wherein the set of operations further comprises: when it is determined that the image data is not authentic, performing additional processing of the image data (Viktorovich, para. [0068]-[0069]; FIG. 1; para. [0058]-[0060]: “In subprocess 870, the JPEG image may be determined to be falsified. In this case, the output of process 800 may be an indication that the JPEG image is an authentic image of an identity document. In addition, in subprocess 880, the falsified area(s) of the JPEG image may be localized as described above. In this case, the output of process 800 may comprise an identification of the location of each area that has been falsified and/or an identification of each field in the identity document that has been falsified.”; see steps 870 and 880 in FIG. 8 in the rejection of claim 1 above; “Localization of Falsified Areas; An auxiliary image I4 is formed from the image I3, on which morphological disconnection and blurring are additionally applied to eliminate parasitic elements resulting from random deviations of the DCT coefficients. Based on image I4, the maximum is calculated—an estimate of manipulations probability. Image I4 is used to construct image I5 with the sizes of the original image. The pixel value within each 8×8 block of image I5 is equal to the pixel value of image I4 corresponding to this block. Estimation of manipulations probability and image I5 are the result of the algorithm operation. For this example (FIG. 2), the estimate of manipulations probability is 0.68302. Contrasted images I4 and I5 are shown in FIGS. 6 and 7, respectively. thus, the above method for detecting and localizing a falsified area in JPEG images makes it possible to detect some cases of falsification of data in an image without reference to its content and has the potential to be used to verify the authenticity of images of documents of various types.”).
With regards to independent claim 14, and dependent claims 15, 17, and 19, they recite the functions of the apparatuses of claims 1-2, 4, and 6, respectively, as processes. Thus, the analyses in rejecting claims 1-2, 4, and 6 are equally applicable to claim 14-15, 17, and 19, respectively.
Claims 3, 7, 10, 13, 16, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Viktorovich, in view of Gueguen, and in further view of U.S. Patent Publication No.: 12,327,402 (Ghosh et al. (hereinafter Ghosh).
Regarding claim 3, Viktorovich, in view of Gueguen, teaches the system of claim 1, wherein generating the set of features for the groups of pixels within each block size comprises: generating a set of discrete cosine transform (DCT) coefficients for the groups of pixels within each block size (Viktorovich, para. [0064]-[0069] teaches generating DCT coefficients for a plurality of blocks having a single size; Gueguen, col. 4, lines 30-64; col. 5, lines 11-16; and FIG. 2 teaches generating DCT coefficients for a plurality of block sizes having different sizes; see rejection of claim 1 above).
Viktorovich, in view of Gueguen, fails to teach
comparing the DCT coefficients to an expected distribution.
Ghosh teaches
comparing the DCT coefficients to an expected distribution (Ghosh, col. 3, lines 65-67; co. 4, lines 1-40: “JPEG images can be compressed according to 8×8 Discrete Cosine Transform (DCT) blocks. The system of the present disclosure can use this fact to detect tampering operations under various principles. JPEG images which are tampered can suffer from a double phenomenon known as double compression, with inconsistencies between DCT histograms of singly and doubly compressed regions. DCT coefficients of unmodified areas can undergo a double JPEG compression thus exhibiting double quantization (DQ) artifacts, while DCT coefficients of tampered areas will result from a single compression and could present no artifacts. The system can identify tampered blocks by considering DCT coefficients and computing a likelihood map indicating the probability for each 8×8 DCT block of being double compressed. Low frequency coefficients within each block can be used in practice to determine the probability that a block is tampered assuming that the DCT coefficients within a block are mutually independent. The likelihoods can be computed based on a Bayesian approach according to the evaluated periodicity of the DCT coefficient histograms. The difference lies in the choice of assumed distribution. Further, the system can identify double JPEG artifacts, including but not limited to A-DJPG and NA-DJPG artifacts. This could depend on whether the second JPEG compression adopts a discrete cosine transform (DCT) grid aligned with the one used by the first compression or not. The system can also identify a specialized feature set to classify blocks using SVM as being double compressed or not. DCT coefficients of single quantized images can follow Benford's distribution, while it may not for double quantized images. The system can address images in the JPEG2000 format and can work where the second quantization factor is small. The system can detect splicing, inpainting, cropping, shifting and other forgery methods. The system can separate two conditional probabilities in making its determinations. Moreover, the system can have direct probabilistic interpretation. For non-aligned double JPEG compression (grid shifted tampering), the system can also employ a threshold-based detector. The system can also employ a threshold-based detector for other situations. Furthermore, for double JPEG compression, the system can generate a likelihood map for tampering.”).
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the set of operations, as taught by Viktorovich, in view of Gueguen, to include the step of comparing the DCT coefficients to an expected distribution, as taught by Ghosh.
The suggestion/motivation for doing so would have been to “detect splicing, inpainting, cropping, shifting and other forgery methods” (Ghosh, col. 3, lines 65-67; co. 4, lines 1-40) in images.
Therefore, it would have been obvious to combine Viktorovich and Gueguen, with Ghosh, to obtain the invention as specified in claim 3.
Regarding claim 7, Viktorovich, in view of Gueguen, teaches the system of claim 1.
Viktorovich, in view of Gueguen, fails to teach
wherein the image data is obtained from a remote computing device.
Ghosh teaches
wherein the image data is obtained from a remote computing device (Ghosh, col. 8, lines 60-67; col. 9, lines 1-5; FIG. 4: “FIG. 4 is a system diagram of an embodiment of a system 30 of the present disclosure … The computer system 32 can also receive wireless or remotely the dataset having the computer images to be processed by the image forgery process of the present disclosure.”;
PNG
media_image9.png
484
698
media_image9.png
Greyscale
).
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the set of operations, as taught by Viktorovich, in view of Gueguen, to include the step of obtaining the image data from a remote computing device, as taught by Ghosh.
The suggestion/motivation for doing so would have been to offload processing to save computing resources for the processing of the images as either authentic or tampered with.
Therefore, it would have been obvious to combine Viktorovich and Gueguen, with Ghosh, to obtain the invention as specified in claim 7.
Regarding claim 10, Viktorovich, in view of Gueguen, teaches the method of claim 8, wherein generating the set of features for the group of pixels within each block size comprises: generating a set of discrete cosine transform (DCT) coefficients for the groups of pixels within each block size (Viktorovich, para. [0064]-[0069] teaches generating DCT coefficients for a plurality of blocks having a single size; Gueguen, col. 4, lines 30-64; col. 5, lines 11-16; and FIG. 2 teaches generating DCT coefficients for a plurality of block sizes having different sizes; see rejection of claim 8 above).
Viktorovich, in view of Gueguen, fails to teach
comparing the DCT coefficients to an expected distribution.
Ghosh teaches
comparing the DCT coefficients to an expected distribution (Ghosh, col. 3, lines 65-67; co. 4, lines 1-40: “JPEG images can be compressed according to 8×8 Discrete Cosine Transform (DCT) blocks. The system of the present disclosure can use this fact to detect tampering operations under various principles. JPEG images which are tampered can suffer from a double phenomenon known as double compression, with inconsistencies between DCT histograms of singly and doubly compressed regions. DCT coefficients of unmodified areas can undergo a double JPEG compression thus exhibiting double quantization (DQ) artifacts, while DCT coefficients of tampered areas will result from a single compression and could present no artifacts. The system can identify tampered blocks by considering DCT coefficients and computing a likelihood map indicating the probability for each 8×8 DCT block of being double compressed. Low frequency coefficients within each block can be used in practice to determine the probability that a block is tampered assuming that the DCT coefficients within a block are mutually independent. The likelihoods can be computed based on a Bayesian approach according to the evaluated periodicity of the DCT coefficient histograms. The difference lies in the choice of assumed distribution. Further, the system can identify double JPEG artifacts, including but not limited to A-DJPG and NA-DJPG artifacts. This could depend on whether the second JPEG compression adopts a discrete cosine transform (DCT) grid aligned with the one used by the first compression or not. The system can also identify a specialized feature set to classify blocks using SVM as being double compressed or not. DCT coefficients of single quantized images can follow Benford's distribution, while it may not for double quantized images. The system can address images in the JPEG2000 format and can work where the second quantization factor is small. The system can detect splicing, inpainting, cropping, shifting and other forgery methods. The system can separate two conditional probabilities in making its determinations. Moreover, the system can have direct probabilistic interpretation. For non-aligned double JPEG compression (grid shifted tampering), the system can also employ a threshold-based detector. The system can also employ a threshold-based detector for other situations. Furthermore, for double JPEG compression, the system can generate a likelihood map for tampering.”).
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the method, as taught by Viktorovich, in view of Gueguen, to include the step of comparing the DCT coefficients to an expected distribution, as taught by Ghosh.
The suggestion/motivation for doing so would have been to “detect splicing, inpainting, cropping, shifting and other forgery methods” (Ghosh, col. 3, lines 65-67; co. 4, lines 1-40) in images.
Therefore, it would have been obvious to combine Viktorovich and Gueguen, with Ghosh, to obtain the invention as specified in claim 10.
Regarding claim 13, Viktorovich, in view of Gueguen, teaches the method of claim 8.
Viktorovich, in view of Gueguen, fails to teach
wherein obtaining the image data comprises obtaining the image data from a remote computing device.
Ghosh teaches
wherein obtaining the image data comprises obtaining the image data from a remote computing device (Ghosh, col. 8, lines 60-67; col. 9, lines 1-5; FIG. 4: “FIG. 4 is a system diagram of an embodiment of a system 30 of the present disclosure … The computer system 32 can also receive wireless or remotely the dataset having the computer images to be processed by the image forgery process of the present disclosure.”;
PNG
media_image9.png
484
698
media_image9.png
Greyscale
).
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify method, as taught by Viktorovich, in view of Gueguen, to include the step of obtaining the image data from a remote computing device, as taught by Ghosh.
The suggestion/motivation for doing so would have been to offload processing to save computing resources for the processing of the images as either authentic or tampered with.
Therefore, it would have been obvious to combine Viktorovich and Gueguen, with Ghosh, to obtain the invention as specified in claim 13.
With regards to claims 16 and 20, they recite the functions of the apparatus of claims 3 and 7, respectively, as processes. Thus, the analyses in rejecting claims 3 and 7 are equally applicable to claims 16 and 20, respectively.
Claims 5 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Viktorovich, in view of Gueguen, and in further view of U.S. Patent Application Publication No.: 2019/0266614 (Grandhi et al. (hereinafter Grandhi).
Regarding claim 5, Viktorovich, in view of Gueguen, teaches the system of claim 1.
Viktorovich, in view of Gueguen, fails to teach
wherein: the image data is obtained as a request that is received via an application programming interface (API); and the positive verification indication is provided, via the API, in response to the request.
Grandhi teaches
wherein: the image data is obtained as a request that is received via an application programming interface (API); and the positive verification indication is provided, via the API, in response to the request (Grandhi, para. [0046]; para. [0050]; para. [0126]; para. [0015]: “An example design of a system 100, similar to system 10, is illustrated in FIG. 3A having three layers: user layer 102, system layer 104, and image API layer 106 … Image API layer 106 includes image services 120.”; “User layer 102 also communicates with image API layer 106 by sending an HTTP request; an XML response is then sent from image API layer 106 to user layer 102. When user layer 102 sends a request for image processing directly to image API layer 106, the request is sent in a format specific to the API of image API layer 106. In this way, image services 120 is able to understand the requested service and is able to retrieve the image data on which the service is to be performed”; “FIG. 12 illustrates a module 900 for fraud detection using image-based information. Product information, such as characteristics or other information either received from the input image data or from user input or selection, is provided to fraud detection unit 920. On receipt of the product information, fraud detection unit 920 determines appropriate information for comparison, and accesses product information in product database 922 … Fraud detection unit 920 matches the received information to information stored in product database 922, and where the received information is consistent with the stored information, a confirmation is provided indicating valid images and product information. If, however, the information is not consistent, then a fraud alert is provided.”; “Further example embodiments may include the ability to work with live auctions when searching for similar images. In this way, an image API may operate in real time monitoring a user's activity, so as to include the user's activity in at least one of the image processing steps.”).
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the steps of obtaining image data and providing a positive verification indication for obtained image data, as both taught by Viktorovich, in view of Gueguen, to include the image data being obtained as a request that is received via an application programming interface (API), and the positive verification indication being provided, via the API, in response to the request, respectively, as taught by Grandhi.
The suggestion/motivation for doing so would have been that APIs enable seamless communication and data exchange between disparate software applications, systems, and platforms.; this fosters improved collaboration, streamlines workflows, and creates a more cohesive digital ecosystem; further API’s power richer and more integrated user experiences by enabling applications to access diverse functionalities and data sources, resulting in more comprehensive and intuitive user interfaces; these benefits will improve an image authentication system that can be easily used by user’s to validate images they encounter on the internet.
Therefore, it would have been obvious to combine Viktorovich and Gueguen, with Grandhi, to obtain the invention as specified in claim 5.
With regards to claim 18, it recites the functions of the apparatus of claim 5 as a process. Thus, the analysis in rejecting claim 5 is equally applicable to claim 20.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MICHAEL ADAM SHARIFF whose telephone number is 571-272-9741. The examiner can normally be reached M-F 8:30-5PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Sumati Lefkowitz can be reached on 571-272-3638. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MICHAEL ADAM SHARIFF/
Examiner, Art Unit 2672
/SUMATI LEFKOWITZ/Supervisory Patent Examiner, Art Unit 2672