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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 03 June 2026 has been entered.
Status of Claims
This is a non-final office action in response to the request for continues examination filed 03 June 2026. Claims 1 and 10-11 have been amended. Claims 5, 7-9 and 14-20 have been canceled. Claims 1-4, 6, and 10-13 remain pending and have been examined.
Response to Amendment
Applicant’s amendment to claims 1 and 10-11 has been entered.
Applicant’s amendment is insufficient to overcome the pending 35 U.S.C. 101 rejection. The rejection remains pending and is updated below, as necessitated by amendment.
Applicant’s amendment is sufficient to overcome the pending 35 U.S.C. 103 rejection. The rejection is respectfully withdrawn.
Response to Arguments
Applicant’s arguments regarding the 35 U.S.C. 101 rejection have been fully considered, but are not persuasive. Applicant asserts that the amended claims are not directed to “collecting and storing data for user access,” but instead recite a specific industrial inspection and welding-control architecture in which machine-learning based defect inference data is utilized in a practical application associated with a welding operation in a manner that satisfies at least Step 2A, Prong Two because claim 1 now additionally recites: “allow the client’s terminal to access the defect interference data registered in the first database via the Internet and control a welding machine to perform a welding process based on the accessed defect inference data.” Applicant additionally asserts that the recited feedback architecture enables the learning model to iteratively improve defect inference accuracy using verified inspection results generated during actual welding operations, that in combination with the welding process based on the accessed defect inference data, integrate any alleged abstract idea into a practical application under Step 2A, Prong Two and Step 2B. Examiner respectfully disagrees.
The claim invention recites an improvement in weld machine operation data analysis for nondestructive inspection, not an improvement to technology or a technical field. While the claims include additional elements in the form of a trained machine learning model that is retrained using defect judgment data, requirements that a machine learning model be iteratively trained or dynamically adjusted do not represent a technological improvement because iterative training using selected training material and dynamic adjustments based on data comparison difference is incident to the very nature of machine learning. Determining to retrain a machine learning model using a new training data set when, in essence, it appears that the quality of the old model requires an adjustment, is not itself a technological improvement. Making the determination of whether to re-train a machine learning model is tantamount to applying a mathematical formula and/or performing steps that could be performed in the human mind (data comparison/quality analysis) because a human could determine that the defect inference data and the defect judgment data are different (defect judgement analysis). Per paragraph [0132] of the Specification the feedback is provided by a human: “The evaluation item related to the feedback is based on a positive or negative feedback input by the client 40 in an input field provided for allowing input of a feedback to the judge 50 (for example, a comment field or five-level evaluation) in the judgment result confirmation screen 45, for example.” Superficially, the instant claims may appear to be analogous to those of Desjardins, because like Desjardins, they involve how machine learning models are trained. However, unlike here, the specification in Desjardins explains that the claimed technique involves training the same machine learning model on multiple tasks in a manner that yields a specific result that uses less storage capacity and reduces system complexity: By training the same machine learning model on multiple tasks as described in this specification, once the model has been trained, the model can be used for each of the multiple tasks with an acceptable level of performance. As a result, systems that need to be able to achieve acceptable performance on multiple tasks can do so while using less of their storage capacity and having reduced system complexity. Therefore, the claimed trained machine learning model does not integrate the recited abstract idea into a practical application and does not amount to significantly more than the recited abstract idea.
Regarding the limitation to “allow the client’s terminal to access the defect interference data registered in the first database via the Internet and control a welding machine to perform a welding process based on the accessed defect inference data,” The “control” of a welding machine is broadly and generically claimed such that it is unclear whether the system of a human using a manual, virtual, or remote control process, is controlling the welding machine to perform a welding process. Per paragraph [0016]: “The welding process is a work process performed by the welder using a welding machine 6 under a welding condition determined for each weld point 11.” The broadest reasonable interpretation of the claim language in view of the Specification is that a human welder controls the welding machine and performs “a welding process” based on the accessed defect inference data. The human interpretation of the results of the claimed data collection and analysis steps, as claimed, is insufficient to confer patent eligibility. As a result, the 35 U.S.C. 101 rejection is proper and maintained.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-4, 6, 8-13, and 20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Independent claim 1 recites a non-destructive inspection judgement data management device and independent claim 11 is recites to a process for non-destructive inspection judgement data management. Independent claims 1 and 11 are recite to an abstract idea of collecting and storing data for user access and analysis. Independent claims 1 and 11 recite substantially similar limitations.
Taking independent claim 11 as representative, claim 11 recites at least the following limitations:
registering image data related to a non-destructive inspection of a weld point in a first database;
registering judge information on a judge undertaking a judgment task of judging whether a weld defect is present at the weld point based on the image data, in a second database;
allowing, based on contract information indicating that a contract for the judgment task has been established with the judge selected as a contractor of the judgment task among the judges registered in the second database, the contractor to access the image data related to the judgment task in the image data registered in the first database via Internet receive defect judgment data indicating a result of judgment of whether the weld defect is present at the weld point based on the image data related to the judgment task from the contractor via the Internet and register the defect judgment data in a third database in association with the weld point;
allowing a client’s terminal of the judgment task to access the defect judgment data related to the judgment task in the defect judgment data registered in the third database via the Internet based on the contract information;
forming a learning data pair including the image data registered in the first database as input data and the defect judgment data registered in the third database for the image data as training data, and
training a learning model based on a plurality of the learning data pairs to cause the learning model to learn a correlation between the input data and the training data by machine learning, and
infer whether the weld defect is present at the weld point as an inference target by inputting the image data of the weld point to the learning model as the input data and register defect inference data indicating a result of inference of whether the weld defect is present at the weld point in the first database in association with the weld point; and in response to the defect inference data and the defect judgment data being different from each other, train the learning model again by using the defect judgment data as ground truth,
wherein in response to a pixel value of each pixel included in the image data being input to an input layer of the learning model, an output layer of the learning model outputs an inference result indicating one or more of presence or absence of the weld defect, a type of the weld defect, and a position of the weld defect as defect inference data; and
allow the client's terminal to access the defect inference data registered in the first database via the Internet and control a welding machine to perform a welding process based on the accessed defect inference data.
Under Step 1, claim 11 recites at least one step or act including allowing the contractor to access the image data related to the judgment task.
Under Step 2A Prong One, the limitations recited in claim 11 for registering image data, registering judge information, allowing the contractor to access the image data, receiving defect judgment data, allowing a client’s terminal of the judgement task to access the defect judgment data, forming a learning pair, training a learning model, inferring whether the weld defect is present at the weld point, training the learning model again, outputting an interference result, and allowing the client’s terminal to access the defect inference data, as drafted, illustrates a process that, under its broadest reasonable interpretation covers performance of the limitation in the mind (collecting and providing access to data), because none of the additional elements preclude the steps from practically being performed in the human mind, or by a human using a pen and paper. Therefore, the limitations fall into the mental processes grouping and accordingly the claims recite an abstract idea.
Under Step 2A Prong Two, claim 11 recites a processor, first, second, and third database for processing data, registering data, and providing access. These elements are recited at a high level of generality (i.e., as a generic data storage elements performing generic data storage functions) and amount to no more than mere instructions to apply the exception using generic computer components. See MPEP 2106.05(f). For example, Applicant’s specification at paragraph [0021] states: “The database device 3 is configured by a general-purpose or dedicated computer (see FIG. 6 described later), for example;” paragraph [0045] further states: “Each of the non-destructive inspection judgment data management device 2, the database device 3, the client's terminal 4, and the judge's terminal 5 is configured by a general-purpose or dedicated computer 900.” Adding generic computer components to perform generic functions, such as data gathering, performing calculations, and outputting a result would not transform the claim into eligible subject matter. See MPEP 2106.05(h). A quality control engineer can visually inspect a weld and mentally determine the presence or absence of a defect, a type of the weld defect, and a position of the weld defect based on past knowledge and experience. Moreover, the claim limitations include information from a human “judge” regarding whether a weld defect is present and storing the defect judgement data in a database are steps for receiving and storing data from the human performance of a task of making mental observations and judgements. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea.
The presence of machine learning algorithm as an additional element and data processing limitations do not necessarily make the claim rooted in computer technology. As recited, the claimed learning model does not improve the functioning of the computing device, it improves the determination of whether a weld defect is present and the type and position of the weld defect if present based on known data or user feedback. While the claims include additional elements in the form of a trained machine learning model that is retrained using defect judgment data, requirements that a machine learning model be iteratively trained or dynamically adjusted do not represent a technological improvement because iterative training using selected training material and dynamic adjustments based on data comparison difference is incident to the very nature of machine learning. Determining to retrain a machine learning model using a new training data set when, in essence, it appears that the quality of the old model requires an adjustment, is not itself a technological improvement. Making the determination of whether to re-train a machine learning model is tantamount to applying a mathematical formula and/or performing steps that could be performed in the human mind (data comparison/quality analysis) because a human could determine that the defect inference data and the defect judgment data are different (defect judgement analysis). Per paragraph [0132] of the Specification the feedback is provided by a human: “The evaluation item related to the feedback is based on a positive or negative feedback input by the client 40 in an input field provided for allowing input of a feedback to the judge 50 (for example, a comment field or five-level evaluation) in the judgment result confirmation screen 45, for example.” Therefore, the claimed trained machine learning model does not integrate the recited abstract idea.
Under Step 2B, independent claim 11 does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of a processor and storage device amount to no more than mere instructions to apply the exception using a generic computer component which cannot provide an inventive concept.
Dependent claims 2-4, 6, 10, and 12-13 include the abstract ideas of the independent claims. The limitations of the dependent claims merely narrow the mental process for managing data by describing how the data is further accessed and analyzed to aid user decision making, without significantly more. The limitations of the dependent claims are not integrated into a practical application because none of the additional elements set forth any limitations that meaningfully limit the abstract idea implementation. There are no additional elements that transform the claim into a patent eligible idea by amounting to significantly more. The analysis above applies to all statutory categories of invention. Accordingly independent claim 1 and the claims that depend therefrom are rejected as ineligible for patenting under 35 U.S.C. 101 based upon the same analysis applied to claim 11 above. Therefore claims 1-4, 6, and 10-13 are ineligible under 35 U.S.C. 101.
Allowable Subject Matter
Claims 1-4, 6, and 10-13 are rejected under 35 U.S.C. 101, but the claims would be allowable if the aforementioned rejections are overcome.
The 35 U.S.C. 103 rejection of the claims is withdrawn in light of Applicant’s Amendments and Remarks filed on 03 June 2026, in particular pg. 14-17, which were deemed persuasive. Examiner analyzed amended Claim 1 and similarly claim 11 in view of the prior art of record and an updated prior art search and finds not all claim limitations are explicitly taught nor would one of ordinary skill in the art find it obvious to combine these references with a reasonable expectation of success as discussed below. Therefore, claims 1-4, 6, and 10-13 are eligible over the prior art.
Metala et al. (US 2009/0307628) discloses hardware and software is illustrated for performing non-destructive testing and evaluation to facilitate inspector interaction with non-destructive examination data, the framework includes a data visualization and analysis application 50. The data visualization and analysis application 50 may be configured to process data extracted from one or more data sources 70, e.g., one or more databases. Messinger et al. (US 2014/0207875) discloses weld inspection, remote visual inspections, and the like, to analyze and detect a variety of conditions, wherein results of the inspection (block 154), may then be analyzed (block 156), for example, by using the NDT device 12, by transmitting inspection data to the cloud 24. The analysis (block 156) may then be reported (block 158), resulting in one or more reports 159, including reports created in or by using the cloud 24 And the application may cross reference the data or type of data indicated to be shared at block 202 with a list of individuals. Perron et al. (US 2022/0244194) discloses visual output 40 can allow a human user to visualize the sequence of acquired images and/or can be used for further determining whether a defect is present in the manufactured article captured in the sequence of acquired images, and a classification module trained by applying a machine learning algorithm to a training captured dataset that includes samples previously presented by the image acquisition device to generate a trained feature set from the training of the classification module from machine learning. However, none of the above listed references teaches the specific ordered sequence of limitations presented in independent claims 1 and 11.
Moreover, since the specific ordered combined sequence of claim elements recited in claims 1 and 11 can only be found as recited in Applicant’s specification, any combination of the cited references and/or additional references to teach all the claim elements, including the features discussed above, would be the result of impermissible hindsight reconstruction. Accordingly, the prior art rejections set forth in the previous action are withdrawn.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Lee (US 2022/0215521) - a non-destructive inspection method based on a transmission image, a method of providing a non-destructive inspection function, and an apparatus for the same, and specifically, when there are transmission images of electronic parts or industrial parts on which an electronic circuit board is mounted, various inspection methods can be used targeting these transmission images so that a user may easily perform defect inspection in accordance with various situations.
Ikeda et al. (US 11,301,978) - the CNN engine can extract features falling within a width of feature quantities defined by the internal parameters. However, a range of feature quantities corresponding to the internal parameters is determined depending on the type of defect included in learning data used for generating a learning-completed model. Thus, when a defect having a peculiar feature not included in the learning data occurs in the production line, there are cases in which a feature quantity of the defect deviates from a feature quantity acquired through the advance learning, and erroneous recognition (overlooking) occurs. In addition, when a pattern of a background area having a peculiar feature not included in the learning data in the production line occurs, there are cases in which the pattern of the background area coincides with the range of feature quantities acquired through the advance learning, and erroneous recognition (excessive extraction) occurs.
Padfield et al. (US 10,769,766) - using artificial intelligence or machine learning systems to automatically determine which of a pre-defined set of visual features are depicted in an image. These visual features are referred to as the “classes” that the models are trained to recognize. The disclosed training techniques overcome the challenge of partial information by labeling known (user-identified) defects as positive cases, not labeling other defects (e.g., those for which no user feedback has been received) as negative cases, and instead using a new label value for defects with an unknown ground truth. This avoids mislabeling a class as not present, when in fact the class may be present but was just unnoticed or unmarked by the user. In addition, the disclosed training techniques use a novel loss function that accounts for this unknown nature of some classes when it determines how to update the machine learning model's parameters.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to LETORIA G KNIGHT whose telephone number is (571)270-0485. The examiner can normally be reached M-F 9am-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, Rutao WU can be reached at 571-272-6045. 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.
/L.G.K/Examiner, Art Unit 3623 /RUTAO WU/Supervisory Patent Examiner, Art Unit 3623
Prosecution Insights