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 Status
The status of claims 1-10 is:
Claims 1-10 were pending as of the Non-Final Rejection mailed 01/15/2026.
Claims 1 and 10 are amended as of the remarks and amendments received 04/14/2026.
Claims 2-9 remain as originally presented as of the remarks and amendments received 04/14/2026.
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 03/26/2026 and 06/15/2026 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements are being considered by the examiner.
Claim Interpretation
The 112(f) interpretations have been withdrawn in light of the amendments received 04/14/2026.
Claim Rejections - 35 USC § 103
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.
Claim(s) 1-3, 5-6, and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Nogami et al. (U.S. Patent Publication No 2020/0118263, hereinafter “Nogami”) in view of Tai et al. (U.S. Patent Publication No 2024/0062525, hereinafter “Tai”).
Regarding claim 1, Nogami discloses an image processing apparatus that estimates each region type of plural types of regions included in an image (Nogami Abstract: “A defect detecting unit detects a defect of an object in an input image”, defect region and non-defect region), comprising a computer (Nogami [0044]: “The functions of the information processing device 100 are realized by a computer, which includes a processor and memory”), wherein a processor of said computer (Nogami [0044]: “The functions of the information processing device 100 are realized by a computer, which includes a processor and memory”) is configured to function as:
a feature extractor configured to output a feature vector corresponding to each of pixels of an image that is input, as intermediate output data (Nogami [0120]: “In a third embodiment, the defect feature amount is extracted from a feature map generated using a CNN (Convolutional Neural Network)”); and
an identifier configured to output output data in which region types of the respective pixels of the image are estimated, on the basis of the intermediate output data output from the feature extractor (Nogami [0127]: “Although the defect determination unit 112 is described here as generating the score map using a sigmoid function, which is often used at the output layer of a neural network, the score map can be generated through a different method instead. For example, the defect determination unit 112 may calculate the score for each pixel by inputting the feature vector v.sub.i,j of each pixel into an SVM”; Nogami [0128]: “Using the score map calculated by the defect determination unit 112, the defect separating unit 113 and the region setting unit 114 can carry out the process of specifying one defect at a time and the process of setting the ROI for each defect, in the same manner as in the first embodiment”), wherein
the region types include at least two types of a first region and a second region (Nogami [0127]: “Although the defect determination unit 112 is described here as generating the score map using a sigmoid function, which is often used at the output layer of a neural network, the score map can be generated through a different method instead. For example, the defect determination unit 112 may calculate the score for each pixel by inputting the feature vector v.sub.i,j of each pixel into an SVM”, defect region and non-defect region),
a region where the closest known feature vector is a first feature vector corresponding to a known pixel belonging to the first region is a domain of the first feature vector in a feature space (Nogami [0127]: “Although the defect determination unit 112 is described here as generating the score map using a sigmoid function, which is often used at the output layer of a neural network, the score map can be generated through a different method instead. For example, the defect determination unit 112 may calculate the score for each pixel by inputting the feature vector v.sub.i,j of each pixel into an SVM”),
a region where the closest known feature vector is a second feature vector corresponding to a known pixel belonging to the second region is a domain of the second feature vector in the feature space (Nogami [0127]: “Although the defect determination unit 112 is described here as generating the score map using a sigmoid function, which is often used at the output layer of a neural network, the score map can be generated through a different method instead. For example, the defect determination unit 112 may calculate the score for each pixel by inputting the feature vector v.sub.i,j of each pixel into an SVM”),
when the feature vector output from the feature extractor belongs to the domain of the first feature vector, the identifier estimates that a pixel corresponding to the feature vector belongs to the first region (Nogami [0128]: “Using the score map calculated by the defect determination unit 112, the defect separating unit 113 and the region setting unit 114 can carry out the process of specifying one defect at a time and the process of setting the ROI for each defect, in the same manner as in the first embodiment”),
and when the feature vector output from the feature extractor belongs to the domain of the second feature vector, the identifier estimates that a pixel corresponding to the feature vector belongs to the second region (Nogami [0128]: “Using the score map calculated by the defect determination unit 112, the defect separating unit 113 and the region setting unit 114 can carry out the process of specifying one defect at a time and the process of setting the ROI for each defect, in the same manner as in the first embodiment”).
Nogami does not explicitly disclose the apparatus, wherein the identifier is updatable for the domain of the first feature vector and the domain of the second feature vector.
However, Tai teaches the apparatus, wherein the identifier is updatable for the domain of the first feature vector and the domain of the second feature vector (Tai [0062]: “According to the first example embodiment, as mentioned above, the updating section 14 updates at least one of the feature extraction section 11, the angle conversion section 12, and the class prediction section 13 with reference to the grouping loss in addition to the source domain classification loss, the target domain classification loss and the conversion loss. Therefore, according to the first example embodiment, knowledge obtained from the source domain is to be used also for training in the target domain. Thus, according to the first example embodiment, the feature extraction section 11 and the class prediction section 12 can be trained even in a case where a small amount of labeled data for target domain is available”).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to the updating of the domains as taught by Tai with the method of Nogami because updating the domains will allow for more accurate domain classifications and because it allows the apparatus to be trained even when there is a small amount of labeled data to train on (Tai [0062]). This motivation for the combination of Nogami and Tai is supported by KSR exemplary rationale (G) Some teaching, suggestion, or modification in the prior art that would have led one of ordinary skill to modify the prior art reference or to combine prior art reference teachings to arrive at the claimed invention and rationale (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results.
Regarding claim 10, it is rejected under the same analysis as claim 10 above.
Regarding claim 2, Nogami discloses the apparatus, wherein the feature space is three-dimensional (Nogami Fig. 11: shows the feature space is three-dimensional).
Regarding claim 3, Nogami discloses the apparatus, wherein the feature extractor is a machine learning model (Nogami [0120]: “In a third embodiment, the defect feature amount is extracted from a feature map generated using a CNN (Convolutional Neural Network)”).
Regarding claim 5, Nogami discloses the apparatus, wherein the image is a captured image of an object, and the first region is a defect region indicating a defect of the object (Nogami [0128]: “Using the score map calculated by the defect determination unit 112, the defect separating unit 113 and the region setting unit 114 can carry out the process of specifying one defect at a time and the process of setting the ROI for each defect, in the same manner as in the first embodiment”).
Regarding claim 6, Nogami discloses a learning method of the extractor in the apparatus, comprising the steps of:
preparing a learning image including the first region and the second region (Nogami [0123]: “On the other hand, the process for detecting the defect can be carried out in the same manner as in the first embodiment, and a CNN can be used to improve the efficiency of the processing”; Nogami [0090]: “Such training data can be prepared as follows. First, a human views the crack image indicated in FIG. 7A, and enters information of the position and width of the crack. FIG. 7B is a diagram illustrating a method for entering this information. As illustrated in FIG. 7B, the human enters the position of the crack indicated in FIG. 7A. For example, the creator of the data can specify the pixels at which a single crack is located, and can specify the pixels at which a different crack is located”);
defining the first region that is known and the second region that is known in the learning image (Nogami [0123]: “On the other hand, the process for detecting the defect can be carried out in the same manner as in the first embodiment, and a CNN can be used to improve the efficiency of the processing”; Nogami [0090]: “Such training data can be prepared as follows. First, a human views the crack image indicated in FIG. 7A, and enters information of the position and width of the crack. FIG. 7B is a diagram illustrating a method for entering this information. As illustrated in FIG. 7B, the human enters the position of the crack indicated in FIG. 7A. For example, the creator of the data can specify the pixels at which a single crack is located, and can specify the pixels at which a different crack is located”);
inputting the first region that is known and the second region that is known, to the feature extractor, and outputting the feature vector from the feature extractor (Nogami [0091]: “Then, on the basis of this data, a set including an image feature amount and a class label is prepared for a single crack, and the classifier F is trained using this set so as to determine a crack width. The image feature amount for the single crack can be extracted in the same manner as in steps S205 and S206”); and
adjusting a parameter of the feature extractor such that the feature vector corresponding to the first region and the feature vector corresponding to the second region are separated from each other in the feature space (Nogami [0107]: “Additionally, in this case, the accuracy at which the attributes are determined can be improved by changing the parameters x (FIG. 6 or FIG. 7B), which are used to determine the range of the ROI, in accordance with the resolution of the input image. In the above-described example, the parameters x express the number of pixels with the defect on their center, and thus it is necessary to change the parameters x in accordance with the resolution in order to use the same part of the detection target as the ROI when the resolution of the input image changes”).
Claim(s) 4 is rejected under 35 U.S.C. 103 as being unpatentable over the Nogami and Tai combination in view of Mendes Rodrigues et al. (U.S. Patent Publication No 2019/0188855, hereinafter “Mendes”).
Regarding claim 4, the Nogami and Tai combination does not explicitly disclose the apparatus, wherein the identifier is a nearest neighbor identifier. (However Nogami does disclose that [0127]: “Although the defect determination unit 112 is described here as generating the score map using a sigmoid function, which is often used at the output layer of a neural network, the score map can be generated through a different method instead”).
However, Mendes teaches the apparatus, wherein the identifier is a nearest neighbor identifier (Mendes [0068]: “For example, a simple classifier such as kNN (k-Nearest Neighbour) could be applied to determine whether a patch from a new scan falls into the defect class or the non-defect class”).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the nearest neighbor identifier as taught by Mendes with the apparatus of the Nogami and Tai combination because it would be a simple substitution (as suggested by Nogami [0127]). This motivation for the combination of Nogami, and Tai, and Mendes is supported by KSR exemplary rationale (B) Simple substitution of one known element for another known element to obtain predictable results.
Claim(s) 7 is rejected under 35 U.S.C. 103 as being unpatentable over the Nogami and Tai combination in view of Theverapperuma et al. (U.S. Patent Publication No 2022/0024485, hereinafter “Theverapperuma”).
Regarding claim 7, the Nogami and Tai combination does not explicitly disclose the method, wherein
the step d) includes adjusting the parameter such that a loss function decreases, and
the loss function is a function in which attractive force acts between feature vectors corresponding to the same region type among the region types and repulsive force acts between feature vectors corresponding to different region types among the region types.
However, Theverapperuma teaches the method, wherein
the step d) includes adjusting the parameter such that a loss function decreases (Theverapperuma [0097]: “The depth images generated as a result of processing the training images can then be compared to corresponding ground truth depth information (e.g., the correct depth value for each pixel in a training image) to adjust the CNN by changing weights and/or bias values for one or more layers of the CNN such that a loss function is minimized”);
and the loss function is a function in which attractive force acts between feature vectors corresponding to the same region type among the region types and the loss function is a function in which attractive force acts between feature vectors corresponding to the same region type among the region types (Theverapperuma [0097]: “The depth images generated as a result of processing the training images can then be compared to corresponding ground truth depth information (e.g., the correct depth value for each pixel in a training image) to adjust the CNN by changing weights and/or bias values for one or more layers of the CNN such that a loss function is minimized”, the loss function is implied to act this way if it is to improve the CNN).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the loss function of Theverapperuma with the method of the Nogami and Tai combination because implementing a loss function would improve the accuracy of the method by further refining the parameters of the extractor which would make its extractions more accurate. This motivation for the combination of Nogami, Tai, and Theverapperuma is supported by KSR exemplary rationale (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results.
Allowable Subject Matter
Claims 8-9 are 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.
Response to Arguments
Regarding Applicant’s arguments on page 9 of Applicant’s arguments and remarks that Nogami’s score map does not read on the domains of the current invention, Examiner respectfully disagrees. There are not limitations in the independent claim that require that defects are visually correlated to each other, only that there are at least two types of a regions. As such, the score map of Nogami discloses the limitations of the independent claim.
Applicant’s arguments with respect to the added limitations to claim(s) 1 and 10 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
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 AIDAN KEUP whose telephone number is (703)756-4578. The examiner can normally be reached Monday - Friday 8:00-4:00.
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, Emily Terrell can be reached at (571) 270-3717. 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.
/AIDAN KEUP/ Examiner, Art Unit 2666
/Molly Wilburn/Primary Examiner, Art Unit 2666
Prosecution Insights