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 .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Response to Amendment
Applicant’s arguments and claim amendments, see P. 4, filed 07/13/2026, with respect to claims 1-8 and 15-17 have been fully considered and are persuasive since claims 1-17 are all canceled. The 35 U.S.C. 112(b) rejection of 04/15/2026 has been withdrawn.
Applicant’s arguments and claim amendments, see P. 7 – P. 9, filed 07/13/2026, with respect to claims 1, 4, 9, and 12 have been fully considered and are persuasive since claims 1-17 are all canceled. The 35 U.S.C. 102 rejection of 04/15/2026 has been withdrawn.
Applicant’s arguments and claim amendments, see P. 9 – P. 18, filed 07/13/2026, with respect to claims 2-3, 5-8, 10-11, and 13-17 have been fully considered and are persuasive since claims 1-17 are all canceled. The 35 U.S.C. 103 rejections of 04/15/2026 has been withdrawn.
Applicant’s addition of claims 18-20 and 25-27 have been considered but are moot in view of the new ground(s) of rejection of Gu et. al (Blind Super-Resolution With Iterative Kernel Correction) in view of ROHEDA et al. (US 2025/0037248 A1).
Applicant’s addition of claims 21-22, 24, 28-29, and 31 have been considered but are moot in view of the new ground(s) of rejection of Gu et. al (Blind Super-Resolution With Iterative Kernel Correction) in view of ROHEDA et al. (US 2025/0037248 A1) and Wang et al. (Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform).
Applicant’s addition and arguments, see P. 2 - P. 18, filed 07/13/2026, with respect to claims 23, 30, and 32-37 have been fully considered and are persuasive. Claims 23, 30, and 32 are objected for dependent upon rejected claims. Claims 33-37 are allowable.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 18-20 and 25-27 are rejected under 35 U.S.C. 103 as being unpatentable over Gu et. al (Blind Super-Resolution With Iterative Kernel Correction, hereinafter Gu) in view of ROHEDA et al. (US 2025/0037248 A1, hereinafter Roheda).
Regarding claims 18 and 25, Gu discloses
Claim 18: An image processing apparatus, comprising: a memory that stores definition data and machine learning parameters for a first machine learning model and a second machine learning model; and a computer processor communicatively coupled to the memory and configured to:
Claim 25: A method, comprising:
obtain target image data (P. 4 Section 3.3: “Suppose the LR image
I
L
R
is of size
C
×
H
×
W
, where
C
denotes the number of channels,
H
and
W
denote the height and width of the image”);
determine, by processing the target image data or a cropped portion thereof with the first machine learning model, at least one tag value (Algorithm line 1, P. 4 Section 3.3: “At the start of the algorithm, an initial estimation
h
0
is given by the predictor function
h
0
=
P
(
I
L
R
)
”),
generate, by processing the image target data and the at least one tag value with the second machine learning model, output image data (Algorithm line 2, P. 4 Section 3.3: “then used to get the first SR result
I
0
S
R
=
F
(
I
L
R
,
h
0
)
”).
However, Gu does not explicitly disclose
wherein each of the at least one tag values corresponds to a type of blur and indicates a degree of the respective type of blur.
Roheda teaches
wherein each of the at least one tag values corresponds to a type of blur and indicates a degree of the respective type of blur (Para [0095]: “The blur detector (204) is further configured to generate a tag comprising an image quality parameter including at least one of the type of blur and the strength of the type of blur, and storing the tag associated with the input image in a media database.”, Para [0104] - Para [0132]).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Gu with determining a tag for type and strength of blur in an image of Roheda, which is in the same field of endeavor of image enhancement, to effectively increase the efficiency and robustness when de-blurring images.
Regarding claims 19 and 26, dependent upon claims 18 and 25 respectively, Gu in view of Roheda teaches everything regarding claims 18 and 25.
Gu further discloses
the at least one tag value comprise a reduction ratio tag, a Gaussian blur tag, a noise tag, and a JPEG compression tag (P. 5 Section 4.1: “We synthesize the training image pairs according to the problem formulation described in section 3.1. For the isotropic Gaussian blur kernels used for training, the kernel width ranges are set to
0.2
,
2.0
,
0.2
,
3.0
a
n
d
[
0.2
,
4.0
]
for SR factors 2, 3 and 4, respectively. We uniformly sample the kernel width in the above ranges. The kernel size is fixed
21
×
21
. to When applying on real world images, we use the additive Gaussian noise with
σ
=
15
. We also provide noise-free version for comparison on the synthetic test images. The HR images are collected from DIV2K [1] and Flickr2K [30], then the training set consists of 3450 high-quality 2K images. The training dataset is augmented with random horizontal flips and 90 degree rotations. All models are trained and tested on RGB channels.”; The current claim language of “at least one” is interpreted as only one of the tag needs to be possible).
Regarding claims 20 and 27, dependent upon claims 18 and 25 respectively, Gu in view of Roheda teaches everything regarding claims 18 and 25.
Roheda further teaches
the first machine learning model comprises a plurality of discriminator models, each discriminator model in the plurality of discriminator models processes, independently, the target image data and outputs a tag value forming the at least one tag value (Para [0104] - Para [0132]; Th global, local, and intentional blur are the plurality of discriminator modes. The C in the equations are determined based on pixels ).
Claim 21-22, 24, 28-29, and 31 are rejected under 35 U.S.C. 103 as being unpatentable over Gu et. al (Blind Super-Resolution With Iterative Kernel Correction, hereinafter Gu) in view of ROHEDA et al. (US 2025/0037248 A1, hereinafter Roheda)and Wang et al. (Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform, hereinafter Wang).
Regarding claims 21 and 28, dependent upon claims 18 and 25 respectively, Gu in view of Roheda teaches everything regarding claims 18 and 25.
However, Gu in view of Roheda does not teach
apply a filter to the target image data forming filter-processed image data; and crop, based on the filter-processed image data, the target image data forming the cropped portion, wherein the first machine learning model processes the cropped portion.
Wang teaches
apply a filter to the target image data forming filter-processed image data; and crop, based on the filter-processed image data, the target image data forming the cropped portion, wherein the first machine learning model processes the cropped portion (Fig. 4, P. 5-6 Section 4: “The mini-batch size was set to 16. The spatial size of cropped HR and LR sub-images were
96
×
96
and
24
×
24
respectively… This is possible since our training images were cropped to contain only one category”; cropping an image result in another image, unless the model is specifically trained to distinguish copped and uncropped images, a cropped image will be interpreted the same as an uncropped image by the model).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Gu in view of Roheda with cropping training images of Wang, which is in the same field of endeavor of image super-resolution, to effectively improve the result of image super-resolution.
Regarding claims 22 and 29, dependent upon claims 21 and 28 respectively, Gu in view of Roheda and Wang teaches everything regarding claims 21 and 28.
Wang further teaches
determining, with the computer processor, a portion of the filter-processed image data that includes a pixel that satisfies a predetermined condition (P. 6 Section 4: “This is possible since our training images were cropped to contain only one category”),
wherein the cropped portion corresponds to the portion of the filter-processed image data (Figure 4, P. 6 Section 4: “For OutdoorSceneTrain, we cropped each image so that only one category exists, resulting in 1k to 2k images for each category.”).
Regarding claims 24 and 31, dependent upon claims 21 and 28 respectively, Gu in view of Roheda and Wang teaches everything regarding claims 21 and 28.
Wang further teaches
dividing, with the computer processor, the filter-processed image into a plurality of blocks (Figure 4, the segmentation); and
determining, with the computer processor, a representative block from that plurality of blocks that includes a pixel that satisfies a predetermined condition, wherein the cropped portion corresponds to the representative block (Section 4: “For OutdoorSceneTrain, we cropped each image so that only one category exists, resulting in 1k to 2k images for each category.”).
Allowable Subject Matter
Claim 23, 30, and 32 objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Claim 33-37 are allowed.
Relevant Prior Art Directed to State of Art
JANG, et al. (WO 2023/013887 A1, hereinafter Jang) is prior art not applied in the rejection(s) above. Jang discloses an electronic device that comprise: a display module; at least one camera disposed on a rear surface of the display module; a memory for storing information related to user authentication; and a processor operatively connected to the display module, the at least one camera, and the memory. The processor can: execute a function for the user authentication; activate the at least one camera in response to the execution; check whether a display area of the display module, corresponding to the position of the at least one camera, is contaminated; and display guide information corresponding to the display area when the contamination of the display area is checked. Various other embodiments may be possible.
CHAE (US 2019/0325557 A1, hereinafter Chae) is prior art not applied in the rejection(s) above. Chae discloses an image processing device that includes: an acquisition unit, a generation unit, a calculation unit, and an estimation unit. The acquisition unit acquires an input image. The generation unit generates a plurality of comparison images by compressing a target region being at least part of the input image with each of a plurality of compression levels and expanding the compressed target region to its original size. The calculation unit calculates, for each of the plurality of comparison images, a degradation level of the comparison image with respect to the input image. The estimation unit estimates the blur level of the input image based on a plurality of calculated degradation levels.
Paola et al. (US 9,934,555 B1, hereinafter Paola) is prior art not applied in the rejection(s) above. Paola discloses a method of processing an image to reduce artifacts, comprising: applying, in a hardware processing unit, an image filter to an image to extract a level of image detail; dividing the image into blocks of a predetermined size; for each block, measuring a density value associated with the level of image detail, wherein measuring the density value associated with the level of image detail comprises using greyscale values associated with the image to calculate a block score indicative of the level of image detail for the block; for blocks wherein the density value exceeds a threshold amount, applying an increasing degree of blur to each block based on an increasing density value, wherein the degree of blur varies between blocks and wherein no blur is applied for blocks having a density value below the threshold amount; and outputting the image such that only some of the blocks of the image are blurred while other blocks do not have blur applied; wherein some of the blurred blocks are not output to the image.
BAIJAL (KR 20210088399 A, hereinafter Baijal) is prior art not applied in the rejection(s) above. Baijal discloses an image display device that includes a display, a memory for storing one or more instructions, and a processor for executing one or more instructions stored in the memory. The processor estimates a blur level for each of a plurality of sub-regions included in a first image by executing one or more instructions, improves the resolution of one or more sub-regions among the plurality of sub-regions based on the estimated blur level. The display can output a second image including one or more sub-regions with improved resolution. A blur in a blurred image can be efficiently removed by using a frequency domain and a spatial domain together.
Lindskog et al. (US 2020/0082535 A1, hereinafter Lindskog) is prior art not applied in the rejection(s) above. Lindskog discloses techniques for the robust usage of semantic segmentation information in image processing techniques, e.g., shallow depth of field (SDOF) renderings.
Goyal et al. (US 9,516,237 B1, hereinafter Goyal) is prior art not applied in the rejection(s) above. Goyal discloses method of calculating blur metrics for a better shutter system when capturing images.
Reddy et al. (US 2020/0242515 A1, hereinafter Reddy) is prior art not applied in the rejection(s) above. Reddy discloses a machine learning model trained using a first set of degraded images for each of a plurality of combinations and corresponding reference images, where a number of degraded images in the first set corresponding to a particular combination of the plurality of combinations is selected in accordance with a probability value associated with the particular combination.
TSAI et al. (US 2023/0021110 A1, hereinafter Tsai) is prior art not applied in the rejection(s) above. Tsai discloses an image segmentation method includes the following steps: obtaining a target image; inputting the target image into a machine learning model to obtain an image segmentation parameter value corresponding to the target image; executing an image segmentation algorithm on the target image according to the image segmentation parameter value to obtain an image segmentation result, wherein the image segmentation result is segmenting the target image into object regions; and displaying the image segmentation result.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSHUA CHEN whose telephone number is (703)756-5394. The examiner can normally be reached M-Th: 9:30 am - 4:30pm ET F: 9:30 am - 2:30pm ET.
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, STEPHEN R KOZIOL can be reached at (408)918-7630. 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.
/J. C./ Examiner, Art Unit 2665
/Stephen R Koziol/ Supervisory Patent Examiner, Art Unit 2665
Prosecution Insights