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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on October 20, 2025 has been entered.
Priority
Acknowledgment is made of applicant's claim for foreign priority based on the Korean application KR10-2020-0058331 and the Korean application KR10-2020-0158043. It is noted, however, that applicant has not filed certified copies of either application as required by 37 CFR 1.55.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1, 7-9 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al. (NPL "Deep Inverse Tone Mapping for Compressed Images") in view of Kim et al. (US 10311337).
Regarding claim 1, Wang discloses a method of extracting noise (page 74559: “the proposed decomposed-based method can remove the compression artifacts”) of a compressed image (page 74559: “compressed LDR image”), the method comprising: obtaining a frequency-band based (Figure 3: “The base layer contains the low frequency components, such as structural and large object information”; the low frequency components represent a frequency band since they span a range of frequencies) first image by applying a convolution filter (page 74562: “use of the guided filter as the filter” where the referenced guided filter is a convolution filter, as evidenced by supporting NPL document “Guided Image Filtering”- see Conclusion) to the compressed image of an original image (page 74562: “One effective method to get the Ibase is filter the input image with an edge-preserving method”); obtaining a second image by subtracting the frequency-band based first image from the compressed image (page 74562: “Then Idetail can be obtained through subtracting the Ibase from the original input image”); and obtaining from the second image, a noise, the noise comprising high-frequency information of the compressed image and a compressed artifact of the compressed image (page 74562: “Idetail is primarily high frequency, i.e. edges, boundaries, noise and compression artifacts”), wherein the frequency-band based first image comprises intermediate-frequency information of the compressed image and low-frequency information of the compressed image (page 74562: “Ibase contains the low frequency components”). However, Wang fails to explicitly disclose the convolution filter sequentially performing down-convolution and up-convolution corresponding to the down-convolution, wherein the convolution filter is trained to obtain, from the compressed image, the frequency-band based first image comprising intermediate-frequency information of the compressed image and low- frequency information of the compressed image by: reducing input feature map of the compressed image having a large dimension to a small dimension input feature map through the down-convolution using a first trained filter kernel to generate a down-convoluted compressed image, and expanding the small dimension input feature map to the large dimension through the up-convolution using a second trained filter kernel to increase the down-convoluted compressed image to an original size of the compressed image. In the related art of deep learning, Kim discloses the convolution filter sequentially performing down-convolution and up-convolution corresponding to the down-convolution (Kim FIG. 7, col 8 lines 25-31: convolution operations and deconvolution operations corresponding to the convolution operations are applied to the input image), wherein the convolution filter is trained (Kim col 10 lines 1-5: “the parameters of the CNN device are optimized through the backpropagation process”) to obtain, from the compressed image, the frequency-band based first image comprising intermediate-frequency information of the compressed image and low- frequency information of the compressed image (Kim FIG. 7, col 8 lines 56-64: “high frequency portions in the size-reduced feature maps is reduced so that the feature maps may have mostly information on low frequency portions. Herein, such low frequency portions mean meaningful parts of the image, e.g., the sky, roads, buildings, automobiles, etc. The result of segmentation of such meaningful parts is obtained through the feature maps outputted through the deconvolution operation, that is, the decoding operation”) by: reducing input feature map of the compressed image having a large dimension to a small dimension input feature map through the down-convolution using a first trained filter kernel to generate a down-convoluted compressed image (Kim col 8 lines 25-28, lines 45-49: “Whenever each convolution operation is applied in the encoding process, the size of the input image is reduced to, for example, ½” where the convolution operation is applied by using a convolution filter), and expanding the small dimension input feature map to the large dimension through the up-convolution using a second trained filter kernel (Kim col 8 lines 28-31: the deconvolution operation is applied by using a deconvolution filter) to increase the down-convoluted compressed image to an original size of the compressed image (Kim FIG. 7: the deconvolution operations increase the size-reduced feature map to the original size since the output segmentation image is the same size as the input image). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Wang to incorporate the teachings of Kim to reduce the amount of operations by reduction of the image size and to extract meaningful parts of the image for segmentation (Kim col 8 lines 48-64).
Regarding claim 7, Wang, modified by Kim, discloses a method of reducing noise (Wang page 74559: “the proposed decomposed-based method can remove the compression artifacts”) of a compressed image (Wang page 74559: “compressed LDR image”), the method comprising: obtaining a frequency-band based first image by applying a convolution filter, the convolution filter sequentially performing down-convolution and up-convolution corresponding to the down-convolution, to the compressed image of an original image; obtaining a second image by subtracting the frequency-band based first image from the compressed image; obtaining from the second image, a noise, the noise comprising high-frequency information of the compressed image and a compressed artifact of the compressed image (as claimed in claim 1); obtaining a third image by removing the compressed artifact by applying a deep neural network (DNN) (Wang Figure 2: Detail Layer Recovery SubNetwork) for removing the compressed artifact, to the noise (Wang Figure 2, page 74558: “The detail layer recovery subnetwork is responsible for the artifacts removal with texture preserving”); and reconstructing the compressed image by summing the frequency-band based first image and the third image (Wang Figure 2, page 74559: “We can obtain an initial result from the sum of these two subnetworks”), wherein the convolution filter is trained to obtain, from the compressed image, the frequency-band based first image comprising intermediate-frequency information of the compressed image and low-frequency information of the compressed image by: reducing input feature map of the compressed image having a large dimension to a small dimension input feature map through the down-convolution using a first trained filter kernel to generate a down-convoluted compressed image, and expanding the small dimension input feature map to the large dimension through the up-convolution using a second trained filter kernel to increase the down-convoluted compressed image to an original size of the compressed image (as claimed in claim 1).
Regarding claim 8, Wang, modified by Kim, discloses the method claimed in claim 7, wherein the DNN for removing the compressed artifact is trained to reduce the compressed artifact (Wang page 74563: “the detail…layer recovery subnetwork will be trained”) with the high-frequency information of the compressed image and the compressed artifact of the compressed image as an input (Wang Figures 2-3: “The detail layer contains the high frequency information, which mainly includes edges, boundaries and compression artifacts”).
Regarding claim 9, Wang, modified by Kim, discloses the method claimed in claim 7, wherein the DNN for removing the compressed artifact comprises a convolution layer (Wang page 74562: The detail layer recovery subnetwork “consists of 2 convolutions layers”) and an activation layer (Wang page 74562: “All of the convolution layers are activated with SELU”).
Regarding claim 15, it is the corresponding apparatus (comprising a memory and a processor) configured to execute the method claimed in claim 7. Therefore, Wang, modified by Kim, discloses the limitations of claim 15 as it does the limitations of claim 7.
Claim(s) 6 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Wang and Kim in view of Schlemper et al. (US 2020/0034998).
Regarding claim 6, Wang, modified by Kim, discloses the method claimed in claim 1. However, Wang fails to disclose the down-convolution and the up-convolution are transposed to each other. In the related art of deep learning, Schlemper discloses the down-convolution (Schlemper Fig. 1A, paragraphs 0091, 0107: “one or more convolutional layers 104”) and the up-convolution are transposed to each other (Schlemper Fig. 1A, paragraphs 0091, 0107: “one or more transposed convolutional layers 108”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have further modified Wang to incorporate the teachings of Schlemper to perform operations at different resolutions to capture both local and global features and eliminate image artifacts (Schlemper paragraph 0107).
Regarding claim 10, Wang, modified by Kim, discloses the method claimed in claim 9. However, Wang fails to disclose the DNN for removing the compressed artifact further comprises a batch normalization layer. In related art, Schlemper discloses the DNN further comprises a batch normalization layer (Schlemper paragraph 0149: “Neural network 520 has a first neural network sub-model 522 with a batch normalization layer 507”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have further modified Wang to incorporate the teachings of Schlemper to improve the performance of the neural network and reduce the time required for training (Schlemper paragraph 0149).
Response to Arguments
Applicant's arguments have been fully considered but they are not persuasive.
Regarding the argument that “the claimed features of a filter trained to separate and obtain low-frequency and intermediate-frequency information according to frequency bands clearly differs in both training objective and criteria from Kim's CNN, which is trained for segmentation and not frequency-based separation”, it is noted that the features upon which the Applicant relies (i.e., a filter trained to separate and obtain low-frequency and intermediate-frequency information according to frequency bands) are not recited in the rejected claim(s). The claims broadly recite the first image is based on frequency bands, not that the first image is obtained by applying a filter trained to separate and obtain low-frequency and intermediate-frequency information according to frequency bands. On the contrary, it is claimed instead that the filter is trained to obtain low-frequency and intermediate-frequency information by performing down-convolution and up-convolution. The frequency-band filter discussed in the specification is separate from the convolution filter performing down-convolution and up-convolution (see FIG. 6). It is unclear how the Applicant is incorporating the features of the frequency-band filter into the convolution filter in their arguments, especially given the lack of support in the specification.
Furthermore, a recitation of the intended use of the claimed invention must result in a structural difference between the claimed invention and the prior art in order to patentably distinguish the claimed invention from the prior art. If the prior art structure is capable of performing the intended use, then it meets the claim. Even though Kim’s CNN is explicitly trained for segmentation, the way the CNN performs segmentation is via frequency separation. The CNN structure performs at least one convolution operation followed by at least one deconvolution operation, trained to produce images with mostly low frequency information containing meaningful parts of the image for segmentation (Kim col 8 lines 40-64). Therefore, Kim’s CNN structure is capable of producing the base layer image in Wang, where the base layer image contains the low frequency components, such as structural and large object information (Wang Figure 3). Additionally, the produced image is frequency-band based since low frequency represents a frequency band, as recited in the above rejection.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
He et al. (NPL “Guided Image Filtering”) discloses the guided filter as a type of explicit image filter, where the filtering process is a translation-variant convolution.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CHRISTINE ZHAO whose telephone number is (703)756-5986. The examiner can normally be reached Monday - Friday 9:00am - 5:00pm EST.
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, Andrew Bee can be reached on (571)270-5183. 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.
/C.Z./Examiner, Art Unit 2677
/Jonathan S Lee/Primary Examiner, Art Unit 2677