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 .
Information Disclosure Statement
The information disclosure statements (IDSs) submitted on May 10, 2024 and August 21, 2025 comply with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements are being considered by the examiner. Citations which have not been considered, have not been considered because they do not comply with 37 CFR 1.98(b) which states “The date of publication supplied must include at least the month and year of publication, except that the year of publication (without the month) will be accepted if the applicant points out in the information disclosure statement that the year of publication is sufficiently earlier than the effective U.S. filing date and any foreign priority date so that the particular month of publication is not in issue”.
35 USC § 101 Statutory Analysis
The claims do not recite any of the judicial exceptions enumerated in the 2019 Revised Patent Subject Matter Eligibility Guidance. Further, the claims do not recite any method of organizing human activity, such as a fundamental economic concept or managing interactions between people. Finally, the claims do not recite a mathematical relationship, formula, or calculation. Thus, the claims are eligible because they do not recite a judicial exception.
Claim Rejections - 35 USC § 102
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 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 the appropriate paragraphs of 35 U.S.C. §102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1-4, 10-13, 15, 16 and 18 are rejected under 35 U.S.C. §102(a)(1) as being anticipated by Li et al. (CN-117710381-A) (hereafter referred to as “Li”).
With regard to claim 1, Li describes obtaining an image of a real-world object (refer for example to page 10, lines 3-4); processing the obtained image using a plurality of vision transformer models to generate a plurality of segmentation masks, each vision transformer model being configured to output a respective segmentation mask corresponding to a respective target material (refer for example to page 2, lines 24-25 and page 16, lines 17-20-which discusses the quality detection and defect analysis of the different materials on the chip, to page 3, lines 13-22, page 10, line 5 through page 11, line 5-which discusses the plurality of vision transformer models, and to page 11, line 6 through page 12, line 2 which discusses the segmentation masks); and using a neural network combiner model, generating a multiphase semantic segmentation mask based on the plurality of segmentation masks, the neural network combiner model trained to integrate outputs of the plurality of vision transformer models, thereby determining semantic segmentation of the real-world object (refer for example to page 2, line 31 through page 3, lines 12-which discusses the neural network combiner model, to page 2, lines 24-25 and page 16, lines 17-20-which discusses the quality detection and defect analysis of the different materials on the chip, to page 3, lines 13-22, page 10, line 5 through page 11, line 5-which discusses the plurality of vision transformer models, and to page 11, line 6 through page 12, line 2 which discusses the segmentation masks, all of which provide for determining semantic segmentation of the real-world object).
As to claim 2, Li describes adapting a given vision transformer model of the plurality of vision transformer models based on a plurality of training data pairs, each of the plurality of training data pairs including a raw training image and an annotated training image, the raw training image and the annotated training image each including indications of at least one target material (refer for example to page 8, lines 3-18).
In regard to claim 3, Li describes wherein the adapting includes configuring the given vision transformer model with one or more of an encoding convolutional layer, a rectified linear unit convolutional layer, and a decoding convolutional layer (refer for example to page 4, lines 7-18, to page 10, line 17 through page 11, line 6, and to page 20, line 25 through page 21, line 14).
With regard to claim 4, Li describes wherein the adapting is configured with at least one of an epoch count between 100 and 50,000, a batch size between 5 and 100, and a learning rate between 0.5 and 0.00001 (refer to page 16, lines 14-17).
With regard to claim 10, Li describes training the neural network combiner model based on a plurality of training data tuples, each of the plurality of training data tuples including a raw training image and respective outputs of the plurality of vision transformer models, the respective outputs being generated by the plurality of vision transformer models based on the raw training image (refer for example to page 13, line 26 through page 14, line 26 and to page 19, line 30 through page 20, line 6).
As to claim 11, Li describes wherein the training is configured with at least one of an epoch count between 50 and 1,000, a batch size between 5 and 20, and a learning rate between 0.5 and 0.00001 (refer for example to page 16, lines 14-17).
In regard to claim 12, Li describes wherein the neural network combiner model includes at least one skip connection between corresponding layers of the neural network combiner model (refer for example to page 13, line 26 through page 14, line 26 and to page 19, line 30 through page 20, line 6).
With regard to claim 13, Li describes wherein the neural network combiner model includes an encoder-decoder module (refer for example to page 4, lines 7-18).
In regard to claim 15, Li describes wherein at least one of the plurality of vision transformer models is a Segment Anything Model (SAM) or a vision transformer Huge model (refer for example to page 20, lines 14-15 and lines 19-22).
With regard to claim 16, Li describes wherein the neural network combiner model is a U-Net model (refer for example to page 17, line 23 through page 18, line 10).
In regard to claim 18, Li describes receiving an indication of interest from a user, the indication of interest corresponding to at least one of a region of the obtained image and one or more materials in the obtained image (refer for example to page , lines ); and wherein generating the plurality of segmentation masks is further based on the received indication of interest (refer for example to page , lines ).
With regard to claim 19, Li describes a processor and a memory with computer code instructions stored thereon, the processor and the memory, with the computer code instructions, being configured to cause the system (refer for example to page 16, lines 11-13) to obtain an image of a real-world object (refer for example to page 10, lines 3-4); process the obtained image using a plurality of vision transformer models to generate a plurality of segmentation masks, wherein each vision transformer model is configured to output a respective segmentation mask corresponding to a respective target material (refer for example to page 2, lines 24-25 and page 16, lines 17-20-which discusses the quality detection and defect analysis of the different materials on the chip, to page 3, lines 13-22, page 10, line 5 through page 11, line 5-which discusses the plurality of vision transformer models, and to page 11, line 6 through page 12, line 2 which discusses the segmentation masks); and using a neural network combiner model, generate a multiphase semantic segmentation mask based on the plurality of segmentation masks, the neural network combiner model trained to integrate outputs of the plurality of vision transformer models, thereby determining semantic segmentation of the real-world object (refer for example to page 2, line 31 through page 3, lines 12-which discusses the neural network combiner model, to page 2, lines 24-25 and page 16, lines 17-20-which discusses the quality detection and defect analysis of the different materials on the chip, to page 3, lines 13-22, page 10, line 5 through page 11, line 5-which discusses the plurality of vision transformer models, and to page 11, line 6 through page 12, line 2 which discusses the segmentation masks, all of which provide for determining semantic segmentation of the real-world object).
As to claim 20, Li describes a non-transitory computer program product for determining semantic segmentation of real-world objects, the computer program product executed by a server in communication across a network with one or more clients and comprising a computer-readable medium, the computer readable medium comprising program instructions, which, when executed by one or more processors, cause the one or more processors (refer for example to page 16, lines 11-13) to obtain an image of a real-world object (refer for example to page 10, lines 3-4); process the obtained image using a plurality of vision transformer models to generate a plurality of segmentation masks, wherein each vision transformer model is configured to output a respective segmentation mask corresponding to a respective target material (refer for example to page 2, lines 24-25 and page 16, lines 17-20-which discusses the quality detection and defect analysis of the different materials on the chip, to page 3, lines 13-22, page 10, line 5 through page 11, line 5-which discusses the plurality of vision transformer models, and to page 11, line 6 through page 12, line 2 which discusses the segmentation masks); and using a neural network combiner model, generate a multiphase semantic segmentation mask based on the plurality of segmentation masks, the neural network combiner model trained to integrate outputs of the plurality of vision transformer models, thereby determining semantic segmentation of the real-world object (refer for example to page 2, line 31 through page 3, lines 12-which discusses the neural network combiner model, to page 2, lines 24-25 and page 16, lines 17-20-which discusses the quality detection and defect analysis of the different materials on the chip, to page 3, lines 13-22, page 10, line 5 through page 11, line 5-which discusses the plurality of vision transformer models, and to page 11, line 6 through page 12, line 2 which discusses the segmentation masks, all of which provide for determining semantic segmentation of the real-world object).
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 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(a) 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 set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied 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.
Claim 14 is rejected under 35 U.S.C. §103(a) as being unpatentable over Li et al. (CN-117710381-A) in view of Bastian et al. (U.S. Patent Application Publication No. US2025/0244261 A1) (hereafter referred to as “Bastian”).
The examiner is providing applicant a copy of PCT/US2023/035542 (filed October 19, 2023) which is the earlier filed related application of U.S. Patent Application Publication No. US2025/0244261 A1.
The arguments advanced in section 6 above, as to the applicability of Li, are incorporated herein.
As to claim 14, although Li does not expressly describe wherein the respective target material is pore, silicon, carbon black-binder or graphite, such materials are well known and widely utilized in the prior art.
Bastian discloses an artificial intelligence anomaly detection using X-ray computed tomography scan data which provides for detecting anomalies in manufactured products using a GAN and CNN (see Figures 15 and 18, and refer for example to paragraphs [0124], [0128], [0144] and [0146]) and describes the respective target material is pore (refer for example to paragraph [0047]).
Given the teachings of the two references and the same environment of operation, namely that of detecting defects in manufactured products, 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 modify the Li system in the manner described by Bastian in order to detect pores according to known methods to yield predictable results and would have been motivated to do so with a reasonable expectation of success in order to provide for increased processing efficiency and higher accuracy as suggested by (refer for example to paragraph [0006]), which fails to patentably distinguish over the prior art absent some novel and unexpected result.
Claim 17 is rejected under 35 U.S.C. §103(a) as being unpatentable over Li et al. (CN-117710381-A) in view of Apte et al. (U.S. Patent Application Publication No. US 2023/0251620 A1) (hereafter referred to as “Apte”).
The arguments advanced in section 6 above, as to the applicability of Li, are incorporated herein.
As to claim 17, although Li does not expressly describe wherein the real-world object is an electrode, such real-world objects are well known and widely utilized in the prior art.
Apte discloses a differentiable model for manufacturability which provides for using CNN models in the manufacturing of semiconductors (see Figure 2 and refer for example to paragraph [0042]) which describes wherein the real-world object is an electrode (refer for example to paragraph [0072]).
Given the teachings of the two references and the same environment of operation, namely that of using CNN models in the manufacturing of semiconductors, 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 modify the Li system in the manner described by Bastian to provide for the real-world object to be an electrode according to known methods to yield predictable results and would have been motivated to do so with a reasonable expectation of success in order to provide for increased processing efficiency and higher accuracy as suggested by Bastian (refer for example to paragraph [0003]), which fails to patentably distinguish over the prior art absent some novel and unexpected result.
Allowable Subject Matter
Claims 5-9 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.
Relevant Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Madjumani, Gu, Schillen and Oh all disclose systems similar to applicant’s claimed invention.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jose L. Couso whose telephone number is (571) 272-7388. The examiner can normally be reached on Monday through Friday from 5:30am to 1:30pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Matthew Bella, can be reached on 571-272-7778. The fax phone number for the organization where this application or proceeding is assigned is (571) 273-8300.
Information regarding the status of an application may be obtained from the Patent Center information webpage on the USPTO website. For more information about the Patent Center, see https://www.uspto.gov/patents/apply/patent-center. Should you have questions about access to the Patent Center, contact the Patent Electronic Business Center (EBC) at 571-272-4100 or via email at: ebc@uspto.gov .
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.
/JOSE L COUSO/Primary Examiner, Art Unit 2667
January 16, 2026