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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 June 23rd, 2026, has been entered.
Response to Amendment
Claims 1-17 in the claim set filed January 26th, 2026, were pending for examination in the Application No. 18/363,088 filed August 1st, 2023. In the remarks and amendments received on June 23rd, 2026, claims 1, 3, 6, 9, 11, 14-15, and 17 are amended, claim 18 is added, and claims 2, 4, 10, and 12 remain canceled. Accordingly, claims 1, 3, 5-9, 11, and 13-18 are currently pending for examination in the application.
Response to Arguments
Applicant’s arguments filed June 23rd, 2026, regarding the rejection(s) of the independent claim(s) have been fully considered but are moot because the arguments do not apply to the new combination of the references being used in the current rejection below.
Priority (Previously Presented)
Acknowledgment is made of applicant’s status as a continuation (CON) of International Application No. PCT/JP2022/005630 filed on February 14th, 2022, which claims priority to provisional U.S. Patent Application No. 63/272,173 filed on October 27th, 2021, and foreign Patent Application No. JP 2021-022117 filed on February 15th, 2021.
Acknowledgment is made of applicant’s claim for foreign priority under 35 U.S.C. 119 (a)-(d). The certified copy has been filed as foreign Patent Application No. JP 2021-022117, filed on October 27th, 2021.
Acknowledgment is made of applicant’s claim for benefit of a prior-filled provisional application under 35 U.S.C. 119(e). The certified copy has been filed as provisional U.S. Patent Application No. 63/272,173, filed on October 27th, 2021.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 18 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Regarding claim(s) 18, the phrase “to such an extent that a region in which a defective portion, shown in the input image, is not changed” modifying the limitation “image style conversion is performed” renders the claim indefinite because it is unclear as to what the metes and bounds satisfies to “such an extent that a region in which a defective portion, shown in the input image is not changed”. The examiner respectfully notes that lines 3-23 of pg. 27 of Applicant’s instant specification details this phrase as the intended use/result of particular structural and/or functional requirements of Applicant’s invention. Therefore, for examination purposes, the phrase “to such an extent that a region in which a defective portion, shown in the input image, is not changed” will be interpreted as reciting intended use/result; and thus, will not be interpreted as a functional or structural requirement of the claim. The examiner respectfully suggests amending into the claims the functional and/or structural requirements that satisfy this phrase as recited in Applicant’s instant specification to overcome this indefiniteness.
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 1, 5-9, and 13-18 are rejected under 35 U.S.C. 103 as being unpatentable over Brauer (US 2022/0036539 A1) in view of Lin et al. (Lin; US 2021/0390682 A1), further in view of Shaham et al. (Shaham; “SinGAN: Learning a Generative Model From a Single Natural Image,” 2019, provided by Applicant’s IDS filed on August 1st, 2023), and further more in view of Riley et al. (US 2019/0294923 A1).
Regarding claim 1, Brauer discloses an image generating method, comprising:
creating a (para(s). [0069] and [0092-0093], recite(s)
[0069] “In general, GANs consist of two adversarial models, a generative model, G, capturing the data distribution, and a discriminative model, D, estimating the probability that a given sample comes from the training data rather than G. …”
[0092] “FIG. 4 illustrates one embodiment of steps that may be performed for artificial image generation using a GAN. In particular, as shown in FIG. 4, the one or more computer subsystems may input design image 400 (conditional image) with added defect 402 into trained generator network 404. The added or synthetic defect may be created in the design data portion shown in design image 400 as described further herein by the one or more computer subsystems. The trained generator network may output generated patch image 406 showing defect 408. In this manner, the synthetic defect is added defect 402 in the portion of design data shown in design image 400, which is input to the GAN by the one or more computer subsystems to thereby generate simulated image 406.”
[0093] “As shown in FIG. 4, therefore, once the GAN is trained, the GAN can be used to generate simulated optical or other images. In this embodiment, the trained generator network is used to create a patch, real-looking image of an artificially introduced defect in a design clip. The generated patch images can then be used as described further herein.”
, where the “trained generator network” is a GAN model (e.g., “generative adversarial networks”) including a first image (e.g., “design image 400”) having a portion of interest shown partially on a target object (e.g., “defect 402”));
extracting(para(s). [0092]—see citation above—, where para(s). [0098] further recite(s)
[0098] “In either case, however, any information for the DOI(s) that are detected on a specimen may be used in step 206 to create design clip(s) that are modified to include the DOI(s). For example, if a bridge type DOI is detected on a specimen, the original design data for the specimen may be modified by the computer subsystem(s) to include bridge type structure(s) at the location at which the bridge type DOI was detected as well as location(s) in the design that are different than the location at which the bridge type DOI was detected. The location(s) at which the DOI is created in a design may be selected based on the types of patterned features located at or near the detected DOI, e.g., so that the modified design clip and the original design clip at the location of a detected DOI contain the same or at least similar patterned structures (similar in shape, dimensions, spacing relative to each other, orientation, etc.). …”
, where obtaining an “added defect 402” is extracting a portion of interest (e.g., “DOI”))
generating an input image by overwriting the extracted portion of interest (para(s). [0092]—see citation in the first limitation of the claim above—, where the “design image 400” is an input image generated by overwriting (e.g., “add[ing]”) the extracted portion of interest (e.g., “synthetic defect”) onto a partial region of a target object image (e.g., “in the portion of design data shown in design image 400”))); and
generating, based on the (para(s). [0092]—see citation above—, where paras. [0089-0090] further recite(s):
[0089] “But in some cases, a first specimen may have similar enough characteristics (e.g., patterned features, materials, etc.) to a second specimen, that a GAN trained on the first specimen can be used to generate simulated images for the second specimen even if the first and second specimens do not have the same designs. …”
[0090] “In this manner, a trained GAN may be repurposed for generating simulated images for specimens it was not necessarily trained for. In one such example, if two different specimens with two different designs have at least some patterned features in common in a portion of the design (e.g., similar memory array areas) formed of similar materials and having the same or similar dimensions, a GAN trained for one of the specimens may be capable of producing simulated images in that portion of the design for another of the specimens. …”
, where the “generated simulated image 406” is a second image exhibiting a portion of interest (e.g., “defect 408”) different in mode (e.g., “different design[s]”) from the portion of interest (e.g., “defect 402”) of the first image), wherein
the generating of the second image includes inputting the input image to the generator(para(s). [0069] and [0092-0093]—see citations in the first limitation of the claim above—, where the “design image” is “input to the GAN… to thereby generated simulated image” is inputting the input image (e.g., “design image”) to the generator (i.e., the “GAN” comprises a “generative model G”)), and
the generator(para(s). [0069] and [0092-0093]—see citations in the first limitation of the claim above—, where the input image (e.g., “design image”) comprising a portion of interest (e.g., “defect 402”) shown partially on a target object is input into the “GAN” model comprising a generator (e.g., “generative model G”) is the generator determined based on a layout (e.g., “data distribution”) of the portion of interest shown in the input image).
Where Brauer does not specifically disclose
creating a SinGAN model including a generator and a discriminatorand
generating, based on the SinGAN model and the input image, a second image…;
Lin teaches in the same field of endeavor of generating defect images
creating a SinGAN model including a generator and a discriminator (para(s). [0042] and [0067], recite(s)
[0042] “… The generative adversarial networks model includes two parts, the generative network model and the discrimination network model. …”
[0067] “In embodiments of the present disclosure, images of various surfaces of the defective article including various defects can be synthesized by acquiring the image of the surface of the good article and the image of the defect using an industrial camera in a fixed optical environment. Specifically, an unconditional generation model that can learn from a single natural image may be used to combine to generate a defect article surface image. The unconditional generation model may include sinGAN, DCGAN, or CGAN, etc. …According to the present embodiment, image samples of different article surfaces containing different defects are acquired as a training set by attaching different defects to the surfaces of different good articles, so that the trained defect detection model has strong practicability.”
, where the unconditional generation model “sinGAN” is a SinGAN model based on a first image (e.g., “image of the defect”) having a portion of interest (e.g., “defect”) shown partially on a target); and
generating, based on the SinGAN model and the input image, a second image… (para(s). [0067]—see citation above—, where the “generat[ed] defect article surface image” is the second image).
Since Brauer and Lin each disclose a GAN model including a generator and a discriminator based on a first image having a portion of interest (e.g., a defect) shown partially on a target object, a person of ordinary skill in the art before the effective filing date of the claimed invention would have recognized that the GAN model of Brauer could have been substituted for the SinGAN model of Lin because both the GAN model and SinGAN model serve the purpose of generating a second image exhibiting a portion of interest different in mode from the portion of interest in the first image.
Where Brauer in view of Lin does not specifically disclose
creating a SinGAN model including a generator and a discriminator in each of a plurality of layers…; …wherein
the generating of the second image includes inputting the input image to the generator in an intermediate layer among the plurality of layers, and
the generator in the intermediate layer is determined based on a layout of the portion of interest shown in the input image;
Shaham teaches in the same field of endeavor of a SinGAN model
creating a SinGAN model including a generator and a discriminator in each of a plurality of layers… (Fig. 4 on pg. 4571, recite(s)
PNG
media_image1.png
568
998
media_image1.png
Greyscale
, where each level of the “pyramid of GANs” is a layer comprising a generator (Gn) and a discriminator (Dn)); …wherein
the generating of the second image includes inputting the input image to the generator in an intermediate layer among the plurality of layers (Fig. 4 on pg. 4571—see citation in claim 1 above—, where the figure depicts the input image (i.e., an image inputted to the SinGAN model, such as
x
^
N
in Fig. 4) is inputted to the generator in an intermediate layer (e.g., “
G
N
-
1
”)), and
the generator in the intermediate layer is determined based on a layout of the portion of interest shown in the input image (Fig. 4 on pg. 4571—see citation in claim 1 above—, where section 2.1. on pg. 4571 further recite(s):
[2.1. Mult-scale architecture] “Our model consists of a pyramid of generators,
{
G
0
,
.
.
.
,
G
N
}
, trained against an image pyramid of
x
:
{
x
0
,
.
.
.
,
x
N
}
, where
x
N
is a downsampled version of
x
by a factor
r
n
, for some
r
>
1
. Each generator
G
N
is responsible of producing realistic image samples w.r.t. the patch distribution in the corresponding image
x
N
. This is achieved through adversarial training, where
G
N
learns to fool an associated discriminator
D
N
, which attempts to distinguish patches in the generated samples from patches in
x
N
.”
, where a person of ordinary skill in the art before the effective filing date of the claimed invention would have recognized that, since the input image comprises the portion-of-interest (e.g., “defect”) and outputs an image comprising the portion-of-interest as disclosed by Brauer in claim 1 above—see citations in claim 1 limitation “generating an input image…” above—, the generator in the intermediate layer which takes in the input image (e.g.,
x
^
N
in Fig. 4) and outputs the same patch distribution in the corresponding input image is determining the generator in the intermediate layer based on at least a layout (e.g., “distribution”) of the portion of interest shown in the input image).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the presently filed invention for the SinGAN model of Brauer in view of Lin to include a generator and a discriminator in each of a plurality of layers comprising of inputting the input image to the generator in an intermediate layer among the plurality of layers, wherein the generator in the intermediate layer is determined based on a layout of the portion of interest shown in the input image, because a SinGAN model inherently comprises of a generator and a discriminator in each of a plurality of layers comprising of generators in intermediate layers determined based on a layout shown in the input image as disclosed by Shaham above.
Where Brauer, as modified by Lin and Shaham, does not specifically disclose
extracting, from an image of interest having the portion of interest shown partially on the target object, only a region in which the portion of interest is shown, as a portion of interest image; and
generating an input image by overwriting the extracted portion of interest image onto a partial region of a target object image excluding the portion-of-interest;
Riley teaches in the same field of endeavor of generating defect images
extracting, from an image of interest having the portion of interest shown partially on the target object, only a region in which the portion of interest is shown, as a portion of interest image (para(s). [0065], recite(s)
[0065] “In another embodiment, the one or more alterations include creating a synthetic defect in the at least one of the one or more images. …Creating the synthetic defect may be performed using one or more of the image editing tools described further herein including, but not limited to, a drawing tool, a painting tool, an erasing tool, cut and paste tools, and the like. For example, the user may manually draw or paint a synthetic defect into an image in the GUI, the user may manually erase a portion of an image in the GUI to thereby create a synthetic defect in the image, and the user may also cut and/or copy a defect image or a portion of an image and paste it into another image to create a synthetic defect in the other image.”
, where “cut[ting] and/or copy[ing] a defect image or a portion of an image and paste it into another image to create a synthetic defect in the other image” is extracting at least a portion of interest image from an image of interest (e.g., a “defect image”) having the portion of interest shown partially on the target object and only show a region in which the portion of interest is shown (i.e., copy “a portion” of a “defect image)); and
generating an input image by overwriting the extracted portion of interest image onto a partial region of a target object image excluding the portion-of-interest (para(s). [0065]—see citation in the preceding limitation immediately above—, where “past[ing]” the extracted portion of interest (e.g., “portion” of a “defect image”) “into another image to create a synthetic defect in the other image” is generating an input image by overwriting the extracted portion of interest image onto a partial region of a target object image excluding the portion-of-interest (e.g., the “other image” that is defect-free)).
Since Brauer also discloses generating a synthetic defect image based on one or more actual defects detected on one or more specimens (Brauer; para. [0102], recite(s)
[0102] “In some such embodiments, the one or more computer subsystems are configured for determining one or more characteristics of the synthetic defect based on one or more defects detected on one or more specimens. …”
), it would have been obvious to one of ordinary skill in the art before the effective filing date of the presently filed invention to further incorporate into the system of Brauer, as modified by Lin and Shaham, generating the input image by overwriting the extracted portion of interest image onto a partial region of a target object image excluding a portion-of-interest image extracted from an image of interest having the portion of interest shown partially on the target object to generate diverse input images for training machine learning models as taught by Riley (para(s). [0086], recite(s)
[0086] “Results produced by training a machine learning model with the synthetic or painted defects described herein have shown that the trained machine learning model can catch real DOIs from synthetic examples even when the fidelity of the synthetic or painted defects to the real DOIs is relatively low. …”
).
Regarding claim 5, Brauer, as modified by Lin, Shaham, and Riley, discloses the image generating method according to claim 1,
wherein Brauer further discloses the generating of the input image includes acquiring region information on the portion of interest (para(s). [0092]—see citation in claim 1 limitation “creating a SinGAN model…” above—, where the “added defect 402” is the acquired region information on the portion of interest (e.g., image information on the defective portion of the input image)), and
wherein the generating of the second image includes:
inputting the input image to the SinGAN model to generate an output image exhibiting the portion of interest different in mode from the portion of interest of the first image (para(s). [0092] and [0089-0090]—see citations in claim 1 limitation “generating, based on the SinGAN model and the input image, a second image…” above—, where the “generated simulated image 406” is an output image exhibiting the portion of interest (e.g., “defect”) different in mode (e.g., “different design[s]”) from the portion of interest (e.g., “defect 402”) of the first image; and Lin teaches the GAN model as a SinGAN model—see teaching of Lin in claim 1 above); and
generating, based on the region information, the second image including the portion of interest included in the output image (para(s). [0092] and [0089-0090]—see citations in claim 1 limitation “generating, based on the SinGAN model and the input image, a second image…” above—, where the “generated simulated image 406” is also a second image included in the output image).
Regarding claim 6, Brauer, as modified by Lin, Shaham, and Riley, discloses the image generating method according to claim 1, wherein Lin further teaches the generating of the second image includes outputting an output image from the SinGAN model (para(s). [0067]—see citation in claim 1 above—, where the “generat[ed] defect article surface image” is an output image from the SinGAN model),
wherein Shaham further teaches the outputting of the output image includes:
inputting a random noise to the generator in at least a lowest layer (Fig. 4 on pg. 4571—see citation in claim 1 above—, where
z
N
in Fig. 4 is random noise inputted to the generator in at least a lowest layer (e.g., generator
G
N
in Fig. 4)); and
outputting the output image including the portion of interest image from the generator in a highest layer (Fig. 4 on pg. 4571—see citation in claim 1 above—, where the output image
x
0
in Fig. 4 is an output image outputted from the generator in the highest layer (e.g., generator
G
0
in Fig. 4); and since the input image comprises the portion-of-interest (e.g., “defect”) and outputs an image comprising the portion-of-interest as disclosed by Brauer in claim 1 above—see citations in claim 1 limitation “generating an input image…” above— and Shaham teaches in section 2.1. on pg. 4571—see citation in claim 4 above—that the generators in a SinGAN model outputs the same patch distribution in the corresponding input image, that a person of ordinary skill in the art before the effective filing date of the claimed invention would have recognized that an image outputted by the generator at the highest layer of the SinGAN model which outputs the same patch distribution in the corresponding input image would output an output image comprising the portion-of-interest (e.g., defect) shown in the input image).
Regarding claim 7, Brauer, as modified by Lin, Shaham, and Riley, discloses the image generating method according to claim 1, wherein Shaham further teaches the generating of the second image includes:
inputting a random noise to the generator in at least a lowest layer (Fig. 4 on pg. 4571—see citation in claim 1 above—, where
z
N
in Fig. 4 is random noise inputted to the generator in at least a lowest layer (e.g., generator
G
N
in Fig. 4)); and
outputting the second image from the generator in a highest layer (Fig. 4 on pg. 4571—see citation in claim 1 above—, where
x
0
in Fig. 4 is the second image outputted from the generator in a highest layer).
Regarding claim 8, Brauer, as modified by Lin, Shaham, and Riley, discloses the image generating method according to claim 1, wherein Brauer further discloses the portion of interest comprises a defective portion shown partially on the target object (para(s). [0092]—see citation in claim 1 limitation “creating a SinGAN model…” above—, where Fig. 4 further depicts:
PNG
media_image2.png
207
551
media_image2.png
Greyscale
, where the “defect 402” is shown partially in “design image 400” as depicted in Fig. 4 above).
Regarding claim 9, the claim differs from claim 1 in that the claim is in the form of an image generating device, comprising: at least one processor; and at least one memory device configured to store a plurality of instructions to be executed by the at least one processor, wherein the at least one memory device is configured to store the SinGAN model of claim 1, and wherein the plurality of instructions cause the at least one processor to execute the method of claim 1. Brauer discloses said processor and memory device (para(s). [0042], recite(s)
[0042] “…In general, the term “computer system” may be broadly defined to encompass any device having one or more processors, which executes instructions from a memory medium. The computer subsystem(s) or system(s) may also include any suitable processor known in the art such as a parallel processor. In addition, the computer subsystem(s) or system(s) may include a computer platform with high speed processing and software, either as a standalone or a networked tool.”
, where the “memory medium” is a memory device). Therefore, claim 9 recites similar limitations to claim 1 and is rejected for similar rationale and reasoning (see the analysis for claim 1 above).
Regarding claim 13, the claim recites similar limitations to claim 5 and is rejected for similar rationale and reasoning (see the analysis for claim 5 above).
Regarding claim 14, the claim recites similar limitations to claim 6 and is rejected for similar rationale and reasoning (see the analysis for claim 6 above).
Regarding claim 15, the claim recites similar limitations to claim 7 and is rejected for similar rationale and reasoning (see the analysis for claim 7 above).
Regarding claim 16, the claim recites similar limitations to claim 8 and is rejected for similar rationale and reasoning (see the analysis for claim 8 above).
Regarding claim 17, the claim differs from claim 1 in that the claim is in the form of a non-transitory computer-readable information storage medium having stored thereon a program executed by a computer, the program causing the computer to operate as an image generating device configured to execute the method of claim 1. Brauer discloses said non-transitory computer-readable information storage medium (para(s). [0145], recite(s)
[0145] “Program instructions 702 implementing methods such as those described herein may be stored on computer-readable medium 700. The computer-readable medium may be a storage medium such as a magnetic or optical disk, a magnetic tape, or any other suitable non-transitory computer-readable medium known in the art.”
). Therefore, claim 17 recites similar limitations to claim 1 and is rejected for similar rationale and reasoning (see the analysis for claim 1 above).
Regarding claim 18, Brauer, as modified by Lin, Shaham, and Riley, discloses the image generating method according to claim 1, wherein Shaham further teaches the generating of the second image includes selecting a generator in an intermediate layer such that image style conversion is performed to such an extent that a region in which a defective portion, shown in the input image, is not changed (section 2.1. on pg. 4571—see citation in claim 1 above—, where the caption of Fig. 11 further recite(s):
[Fig. 11 caption] “Paint-to-Image. We train SinGAN on a target image and inject a downsampled version of the paint into one of the coarse levels at test time. Our generated images preserve the layout and general structure of the clipart while generating realistic texture and fine details that match the training image. Well-known style transfer methods [17, 38] fail in this task.”
, where “preserv[ing] the layout and general structure” of the input image is at least selecting a generator in an intermediate layer such that image style conversion is performed to such an extent that a region shown in the input image is not changed).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the presently filed invention to modify the system of Brauer, as modified by Lin, Shaham, and Riley, to incorporate selecting a generator in an intermediate layer such that image style conversion is performed to such an extent that a region in which a defective portion, shown in the input image, is not changed because Shaham teaches that the generators in SinGAN preserve the layout and general structure of the input image; and since the input image of the system of Brauer, as modified by Lin, Shaham, and Riley, comprises a defective portion region (e.g., see the “portion of interest” recited in claim 1, such as in the claim 1 limitation “creating a SinGAN model…” above), a person of ordinary skill in the art before the effective filing date of the claimed invention would have recognized that the input image comprising a defective portion region input into a selected generator in an intermediate layer of the SinGAN model would not be changed (i.e., preserved) as taught by Shaham above.
Claims 3 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over Brauer, as modified by Lin, Shaham, and Riley, as applied to claims 1 and 9 above, and further more in view of Ki et al. (Ki; US 2020/0380373 A1).
Regarding claim 3, Brauer, as modified by Lin, Shaham, and Riley, discloses the image generating method according to claim 2, wherein Ki teaches in the same field of endeavor of defect image generation the generating of the input image includes generating the input image by cutting out a region of the portion of interest and a periphery of the portion of interest from the target object image and portion-of-interest image (para(s). [0041], [0055], [0067], and [0080], recite(s)
[0041] “The reconstruction algorithm 200 according to the exemplary embodiment of the present disclosure may be an image reconstruction algorithm. Herein, the image reconstruction algorithm may include, for example, a Variational Autoencoder (VAE) and a Generative Model, and particularly, include generative adversarial networks, conditional generative adversarial networks, and the like. The image reconstruction algorithm is merely an example, and the scope of the present disclosure is not limited thereto.”
[0055] “The first, second, and third training of the present disclosure may be performed by using patches obtained by dividing an entire image in a predetermined size, as input, and may be performed by using a patch extracted from a portion corresponding to a defect in the image as input.”
[0067] “The computing device 100 according to the exemplary embodiment of the present disclosure may input defect data 211 of the source domain, to which the first mask is applied, to the generating network 510 (S310). In the case where the defect data is image data, the computing device 100 according to the exemplary embodiment of the present disclosure may progress a training process by inputting a patch extracted from the entire image to the generating network 510 and may also progress a training process by inputting a patch extracted from a defect part of the image to the generating network 510. …”
[0080] “…The image patch 710 may have a size of N pixels × N pixels, and may have a size of 1 pixel. The image patch of the present disclosure may be extracted from the entirety or a part of the data. For example, the image patch of the present disclosure may be extracted mainly based on the defect part in the defect data. The identification network may output final distinguishment information by aggregating the response values for each image patch obtained by comparing the defect data 213 of the source domain, to which the first mask is not applied, and the generated defect data 515 of the source domain, to which the first mask is reconstructed for each image patch, and the scope of the present disclosure is not limited thereto.”
, where the input to an image “generating network” can be an “image patch” extracted from “part of the [entirety of image] data” mainly “based on the defect part in the defect data” is generating the input image by cutting (i.e., “extract[ing]”) a region of the portion of interest (e.g., the “defect”) and a periphery of the portion of interest from an image comprising a target object image and portion-of-interest image (i.e., an “image patch” comprises of both the defect—i.e., the portion-of-interest—and area around the defect—i.e., the periphery of the portion of interest)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the presently filed invention to modify the system of Brauer, as modified by Lin, Shaham, and Riley, to incorporate cutting out a region of the portion of interest and a periphery of the portion of interest from the target object image and portion-of-interest image to improve input image generation by focusing the image generation network on generating an output image comprising the portion of interest (e.g., the defect) as taught by Ki above.
Regarding claim 11, the claim recites similar limitations to claim 3 and is rejected for similar rationale and reasoning (see the analysis for claim 3 above).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JULIA Z YAO whose telephone number is (571)272-2870. The examiner can normally be reached Monday - Friday (8:30AM - 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, 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.
/J.Z.Y./Examiner, Art Unit 2666
/MING Y HON/Primary Examiner, Art Unit 2666
Prosecution Insights