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 .
DETAILED ACTION
1. Claims 1 - 20 are pending. Claim 1 is independent. File date on 7-24-2025.
Double Patenting
2. The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the "right to exclude" granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory obviousness-type double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Omum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); and In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on a nonstatutory double patenting ground provided the conflicting application or patent either is shown to be commonly owned with this application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement.
Effective January 1, 1994, a registered attorney or agent of record may sign a terminal disclaimer. A terminal disclaimer signed by the assignee must fully comply with 37 CFR 3.73(b).
3. Initially it should be noted that the present application is a continuation application of application 18/666299, now patent 12,400,486 having the same inventive entity. The Assignee in both applications is the same. The entire disclosures of the instant application and the patent are identical.
Claims 1 - 20 are rejected under the judicially created doctrine of nonstatutory obviousness type double patenting as being unpatentable over Claims 1 - 20 of U.S. Patent No. 12,400,486. Although the conflicting claims are not identical, they are not patentably distinct from each other.
Claim 1 of the instant application (19/279492) is almost the same as Patent (12,400,486) Claim 1. Claim 1 of the 12,400,486 Patent as shown in the table below contains every element of Claim 1 of the instant application and as such the difference is not enough to distinguish the two claims. Claim 1 of the instant application therefore is not patently distinct from the earlier patent claim and as such are unpatentable over nonstatutory obviousness type double patenting. A later patent/application claim is not patentably distinct from an earlier claim, if the later claim is unpatentable over the earlier claim.
Application 19/279492
Claim 1
Patent (12,400,486)
Claim 1
“monitoring a plurality of the product manufacturing messages”
“monitoring, by a device for processing product manufacturing messages, a plurality of the product manufacturing messages”
“establishing a plurality of product defect analysis tasks at least based on the plurality of the product manufacturing messages”
“establishing, by the device for processing product manufacturing messages, a product defect analysis task queue comprising a plurality of product defect analysis tasks based on the plurality of the product manufacturing messages”
“generating a product defect analysis request message at least based on the plurality of product defect analysis tasks”
“generating, by the device for processing product manufacturing messages, a product defect analysis request message based on the product defect analysis task queue and the plurality of the product manufacturing messages”
“sending the product defect analysis request message to a product manufacturing assisting device, wherein the product defect analysis request message indicates a product defect analysis distributed to the product manufacturing assisting device”
“sending, by the device for processing product manufacturing messages, the product defect analysis request message to a product manufacturing assisting device, wherein the product defect analysis request message indicates a product defect analysis distributed to the product manufacturing assisting device”
“receiving a product defect analysis response message from the product manufacturing assisting device”
“determining whether a defect identification model corresponding to the product type is present” and “receiving a first product defect analysis response message sent by the first product manufacturing assisting device”
Claim Rejections - 35 USC § 103
4. 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.
5. Claims 1-3, 5-17, 19, 20 are rejected under 35 U.S.C. 103 as being unpatentable over Mukundan et al. (US PGPUB No. 20160147883) in view of Tae et al. (US PGPUB No. 20220245402).
Regarding Claim 1, Mukundan discloses a method for processing product manufacturing messages, wherein the method is executed by a device for processing product manufacturing messages, the method comprising:
a) monitoring a plurality of the product manufacturing messages; (Mukundan Fig. 6; ¶ 002, ll 4-8: requirement to effectively and efficiently monitor the production cycle of the product and its manufacturing process, in order to identify the potential failures and associated risks at an early stage and thus achieve a better quality product; ¶ 030, ll 1-7: user-computing device, to determine the defects per unit of the product; the database server generates the data updating session information with the data sources; The data sources update the centralized data stored in the storage unit of the database server)
b) establishing a plurality of product defect analysis tasks at least based on the plurality of the product manufacturing messages; (Mukundan ¶ 031, ll 1-8: an input is received through the user-interface displayed on the display unit; the input corresponds to NPI ID of the product for which the defects per unit needs to be calculated; ¶ 044, ll 1-25: Upon receiving the NPI ID (product ID), sends a query message to the second database server; The query message corresponds to a query to retrieve the product test data from the database server; sends a retrieval session message to the database server; The retrieval session message corresponds to creation of the data retrieval session between the database server and the user-computing device)
c) generating a product defect analysis request message at least based on the plurality of product defect analysis tasks; (Mukundan ¶ 031, ll 1-8: an input is received through the user-interface displayed on the display unit; the input corresponds to NPI ID (product ID) of the product for which the defects per unit needs to be calculated; ¶ 044, ll 1-25: Upon receiving the NPI ID, sends a query message to the second database server; The query message corresponds to a query to retrieve the product test data from the database server; sends a retrieval session message to the database server; The retrieval session message corresponds to creation of the data retrieval session between the database server and the user-computing device; ¶ 045, ll 1-9: After termination of the data updating session, the database server sends a retrieval message to the user-computing device; The retrieval message corresponds to transmitting the product test data of the product from the database server to the user-computing device; Upon receiving the product test data from the database server, the calculation unit of the user-computing device calculates and determines defects per unit of the product (order of tasks to generate a defect determination result); (selected: the order in which the product manufacturing messages are received); (multiple messages transferred between manufacturing system components))
d) sending the product defect analysis request message to a product manufacturing assisting device, wherein the product defect analysis request message indicates a product defect analysis distributed to the product manufacturing assisting device. (Mukundan ¶ 031, ll 1-8: an input is received through the user-interface displayed on the display unit (request message); the input corresponds to NPI ID (product ID) of the product for which the defects per unit needs to be calculated; ¶ 045, ll 1-9: After termination of the data updating session, the database server sends a retrieval message to the user-computing device; The retrieval message corresponds to transmitting the product test data of the product from the database server to the user-computing device; Upon receiving the product test data from the database server, the calculation unit of the user-computing device calculates and determines defects per unit of the product)
Mukundan does not explicitly disclose for e) receiving a product defect analysis response message from the product manufacturing assisting device.
However, Tae discloses:
e) receiving a product defect analysis response message from the product manufacturing assisting device. (Tae ¶ 015, ll 1-6: determination accuracy of a new learning model (response information) may be determined by various factors, wherein the amount of training data, selection of hyper-parameters initially set for learning, selection of a learning model, and the like are important factors in determining the determination accuracy; ¶ 058, ll 1-6: The present disclosure provides an example in which determination type information includes at least one piece of information including: defect type information for the above-described type of defects, product type information about the type of products to be inspected, and part type information about the type of parts to be inspected; ¶ 013, ll 1-8: a format of an image to be used as training data may be different, type of defects may be different depending on the result even with the same defect type, and the result itself revealing a defect may be different, and thus form of the image is different, whereby each learning model should be generated and applied theoretically according to the result of each process and the type of defects (quantity of images); ¶ 021, ll 1-14: model determination system according to the present invention, a candidate model extraction module may extract at least two or more candidate models from among a plurality of learning models stored in a learning-model storage on the basis of determination type information; (storage address))
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for e) receiving a product defect analysis response message from the product manufacturing assisting device as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 2, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 1, wherein, the establishing the plurality of product defect analysis tasks at least based on the plurality of product manufacturing messages comprises: sorting the plurality of the product defect analysis tasks based on any one or more of: the order in which the product manufacturing messages are received, priorities of products, and a product scheduling plan. (Mukundan ¶ 031, ll 1-8: an input is received through the user-interface displayed on the display unit; the input corresponds to NPI ID (product ID) of the product for which the defects per unit needs to be calculated; ¶ 044, ll 1-25: Upon receiving the NPI ID, sends a query message to the second database server; The query message corresponds to a query to retrieve the product test data from the database server; sends a retrieval session message to the database server; The retrieval session message corresponds to creation of the data retrieval session between the database server and the user-computing device; ¶ 045, ll 1-9: After termination of the data updating session, the database server sends a retrieval message to the user-computing device; The retrieval message corresponds to transmitting the product test data of the product from the database server to the user-computing device; Upon receiving the product test data from the database server, the calculation unit of the user-computing device calculates and determines defects per unit of the product (order of tasks to generate a defect determination result); (selected: the order in which the product manufacturing messages are received))
Regarding Claim 3, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 1, wherein the product defect analysis tasks comprise a task of identifying product defect content based on a defect identification model, wherein the product defect content comprises any one or more of: product defect type, product defect location, and product defect size. (Mukundan ¶ 006, ll 14-19: A measure of defects per unit is then determined based on the product test data retrieved from the database server; The defects per unit is displayed through the display unit; (selected: product defect size))
Regarding Claim 5, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 1, including request message. (Mukundan ¶ 031, ll 1-8: an input is received through the user-interface displayed on the display unit; the input corresponds to NPI ID of the product for which the defects per unit needs to be calculated)
Mukundan does not explicitly disclose for a) obtaining a product type and a product defect analysis task type, and for b) generating the product defect analysis request message based on the product type and the product defect analysis task type.
However, Tae discloses wherein, the generating a product defect analysis request message based on the plurality of product defect analysis tasks comprises:
a) obtaining a product type and a product defect analysis task type; b) generating the product defect analysis request message based on the product type and the product defect analysis task type. (Tae ¶ 031, ll 1-7: AI-based vision inspection management system according to the present disclosure may include: a plurality of vision AI clients respectively installed on a plurality of geographically separated product production lines, and configured to use an AI-based learning model to inspect defects; and a vision AI cloud configured to generate a new learning model to be registered in each vision AI client (AI model not present); ¶ 058, ll 1-6: The present disclosure provides an example in which determination type information includes at least one piece of information including: defect type information for the above-described type of defects, product type information about the type of products to be inspected, and part type information about the type of parts to be inspected)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for a) obtaining a product type and a product defect analysis task type, and for b) generating the product defect analysis request message based on the product type and the product defect analysis task type as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 6, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 1, wherein, the sending the product defect analysis request message to a product manufacturing assisting device comprises:
a) performing message format verification on the product defect analysis request message; (Mukundan ¶ 019, ll 1-11: The package manifest of the new branch is updated based on the set of compatible versions; The files that reference the package manifest in the new branch are updated; The new branch is validated based on the updated files; The server publishes a new version of the updated package to the registry in response to the validation being successful)
b) sending the product defect analysis request message after determining that a message format of the product defect analysis request message is qualified. (Mukundan ¶ 031, ll 1-8: an input is received through the user-interface displayed on the display unit (request message); the input corresponds to NPI ID (product ID) of the product for which the defects per unit needs to be calculated; ¶ 019, ll 1-11: The package manifest of the new branch is updated based on the set of compatible versions; The files that reference the package manifest in the new branch are updated; The new branch is validated based on the updated files; The server publishes a new version of the updated package to the registry in response to the validation being successful)
Regarding Claim 7, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 1, wherein the sending the product defect analysis request message to a product manufacturing assisting device further comprises:
b) sending a first product defect analysis request message to the product manufacturing assisting device (Mukundan ¶ 034, ll 1-7: The plurality of training data may be stored after being transmitted from at least one of the plurality of vision AI clients, the pre-training model determination system may determine the pre-training model according to a model generation request from the vision AI cloud that has transmitted the training data, and the cloud control module may register the new learning model in the vision AI cloud)
Mukundan does not explicitly disclose for a) determining whether a defect identification model corresponding to the product type is present, and for b) to configure a training task for the defect identification model in a case where the defect identification model corresponding to the product type is not present, and for c) sending a second product defect analysis request message to the product manufacturing assisting device to detect a product defect by utilizing the defect identification model.
However, Tae discloses for a) determining whether a defect identification model corresponding to the product type is present; (Tae ¶ 017, ll 1-7: AI-based pre-training model determination system and an AI-based vision inspection management system using the same for a product production line, wherein an optimal pre-training model may be used in generating a new learning model by using a previously registered learning model as a pre-training model; (AI model is present, already registered)), and for b) to configure a training task for the defect identification model in a case where the defect identification model corresponding to the product type is not present; (Tae ¶ 031, ll 1-7: AI-based vision inspection management system according to the present disclosure may include: a plurality of vision AI clients respectively installed on a plurality of geographically separated product production lines, and configured to use an AI-based learning model to inspect defects; and a vision AI cloud configured to generate a new learning model to be registered in each vision AI client (AI model not present); ¶ 058, ll 1-6: The present disclosure provides an example in which determination type information includes at least one piece of information including: defect type information for the above-described type of defects, product type information about the type of products to be inspected, and part type information about the type of parts to be inspected; ¶ 013, ll 1-8: a format of an image to be used as training data may be different, type of defects may be different depending on the result even with the same defect type, and the result itself revealing a defect may be different, and thus form of the image is different, whereby each learning model should be generated and applied theoretically according to the result of each process and the type of defects (quantity of images); ¶ 021, ll 1-14: model determination system according to the present invention, a candidate model extraction module may extract at least two or more candidate models from among a plurality of learning models stored in a learning-model storage on the basis of determination type information; (storage address)), and for c) sending a second product defect analysis request message to the product manufacturing assisting device to detect a product defect by utilizing the defect identification model in a case where the defect identification model corresponding to the product type is present. (Tae ¶ 017, ll 1-7: AI-based pre-training model determination system and an AI-based vision inspection management system using the same for a product production line, wherein an optimal pre-training model may be used in generating a new learning model by using a previously registered learning model as a pre-training model; (AI model is present, already registered))
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for a) determining whether a defect identification model corresponding to the product type is present, and for b) to configure a training task for the defect identification model in a case where the defect identification model corresponding to the product type is not present, and for c) sending a second product defect analysis request message to the product manufacturing assisting device to detect a product defect by utilizing the defect identification model as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 8, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 7.
Mukundan does not explicitly disclose for a) determining product type based on plurality of product manufacturing messages, and for b) determining whether defect identification model corresponding to product type is present.
However, Tae discloses wherein the determining whether a defect identification model corresponding to the product type is present comprises:
a) determining the product type based on the plurality of the product manufacturing messages; (Tae ¶ 058, ll 1-6: The present disclosure provides an example in which determination type information includes at least one piece of information including: defect type information for the above-described type of defects, product type information about the type of products to be inspected, and part type information about the type of parts to be inspected) and
b) determining whether the defect identification model corresponding to the product type is present, based on the product type. (Tae ¶ 017, ll 1-7: AI-based pre-training model determination system and an AI-based vision inspection management system using the same for a product production line, wherein an optimal pre-training model may be used in generating a new learning model by using a previously registered learning model as a pre-training model; (AI model is present, already registered); ¶ 058, ll 1-6: The present disclosure provides an example in which determination type information includes at least one piece of information including: defect type information for the above-described type of defects, product type information about the type of products to be inspected, and part type information about the type of parts to be inspected)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for a) determining product type based on plurality of product manufacturing messages, and for b) determining whether defect identification model corresponding to product type is present as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 9, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 7.
Mukundan does not explicitly disclose performing a prejudgment using defect identification model to determine whether defect identification model corresponding to product type is present.
However, Tae discloses wherein, the determining whether a defect identification model corresponding to the product type is present comprises: performing a prejudgment using the defect identification model to determine whether the defect identification model corresponding to the product type is present. (Tae ¶ 017, ll 1-7: AI-based pre-training model determination system and an AI-based vision inspection management system using the same for a product production line, wherein an optimal pre-training model may be used in generating a new learning model by using a previously registered learning model as a pre-training model; (AI model is present, already registered); ¶ 028, ll 1-9: The determination type information may include at least one of defect type information on a type of defects, product type information on a type of products to be inspected, and part type information on a type of parts to be inspected, the learning models may include model information having at least one of the defect type information, the product type information, and the part type information, and the candidate model extraction module may extract the candidate model with reference to the model information)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for performing a prejudgment using defect identification model to determine whether defect identification model corresponding to product type is present as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 10, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 7.
Mukundan does not explicitly disclose for a) determining defect identification model corresponding to product type is present, and for b) determining defect identification model corresponding to product type is not present.
However, Tae discloses wherein, the determining whether a defect identification model corresponding to the product type is present comprises:
a) in response to the product type indicating a known product, determining the defect identification model corresponding to the product type is present; (Tae ¶ 017, ll 1-7: AI-based pre-training model determination system and an AI-based vision inspection management system using the same for a product production line, wherein an optimal pre-training model may be used in generating a new learning model by using a previously registered learning model as a pre-training model; (AI model is present, already registered); ¶ 028, ll 1-9: The determination type information may include at least one of defect type information on a type of defects, product type information on a type of products to be inspected, and part type information on a type of parts to be inspected, the learning models may include model information having at least one of the defect type information, the product type information, and the part type information, and the candidate model extraction module may extract the candidate model with reference to the model information) and
b) in response to the product type indicating an unknown product, determining the defect identification model corresponding to the product type is not present. (Tae ¶ 031, ll 1-7: AI-based vision inspection management system according to the present disclosure may include: a plurality of vision AI clients respectively installed on a plurality of geographically separated product production lines, and configured to use an AI-based learning model to inspect defects; and a vision AI cloud configured to generate a new learning model to be registered in each vision AI client (AI model not present); ¶ 028, ll 1-9: The determination type information may include at least one of defect type information on a type of defects, product type information on a type of products to be inspected, and part type information on a type of parts to be inspected, the learning models may include model information having at least one of the defect type information, the product type information, and the part type information, and the candidate model extraction module may extract the candidate model with reference to the model information; (model not present, model is registered as new model))
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for a) determining defect identification model corresponding to product type is present, and for b) determining defect identification model corresponding to product type is not present as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 11, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 7.
Mukundan does not explicitly disclose in response to the product type being a known product but no product defect content being identified, determining defect identification model corresponding to product type is not present or in response to performance of the defect identification model being insufficient, determining defect identification model corresponding to product type is not present.
However, Tae discloses wherein, the determining whether a defect identification model corresponding to the product type is present comprises:
in response to the product type being a known product but no product defect content being identified based on the defect identification model corresponding to the product type, determining the defect identification model corresponding to the product type is not present; or
in response to performance of the defect identification model being insufficient to satisfy the product defect analysis tasks corresponding to the product type, determining the defect identification model corresponding to the product type is not present. (Tae ¶ 031, ll 1-7: AI-based vision inspection management system according to the present disclosure may include: a plurality of vision AI clients respectively installed on a plurality of geographically separated product production lines, and configured to use an AI-based learning model to inspect defects; and a vision AI cloud configured to generate a new learning model to be registered in each vision AI client (AI model not present); ¶ 028, ll 1-9: The determination type information may include at least one of defect type information on a type of defects, product type information on a type of products to be inspected, and part type information on a type of parts to be inspected, the learning models may include model information having at least one of the defect type information, the product type information, and the part type information, and the candidate model extraction module may extract the candidate model with reference to the model information)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for in response to the product type being a known product but no product defect content being identified, determining defect identification model corresponding to product type is not present or in response to performance of the defect identification model being insufficient, determining defect identification model corresponding to product type is not present as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 12, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 7.
Mukundan does not explicitly disclose product defect analysis response message comprises one or more of: identity, accuracy, and recall of the defect identification model, and wherein defect identification model is determined based on product type, the storage address of the product images, and quantity of the product images.
However, Tae discloses wherein the receiving the product defect analysis response message comprising:
a) receiving the product defect analysis response message from the product manufacturing assisting device, wherein the product defect analysis response message comprises one or more of: identity, accuracy, and recall of the defect identification model; b) wherein the defect identification model is determined based on the product type, the storage address of the product images, and the quantity of the product images. (Tai ¶ 015, ll 1-6: determination accuracy of a new learning model (response information) may be determined by various factors, wherein the amount of training data, selection of hyper-parameters initially set for learning, selection of a learning model, and the like are important factors in determining the determination accuracy; ¶ 058, ll 1-6: The present disclosure provides an example in which determination type information includes at least one piece of information including: defect type information for the above-described type of defects, product type information about the type of products to be inspected, and part type information about the type of parts to be inspected; ¶ 013, ll 1-8: a format of an image to be used as training data may be different, type of defects may be different depending on the result even with the same defect type, and the result itself revealing a defect may be different, and thus form of the image is different, whereby each learning model should be generated and applied theoretically according to the result of each process and the type of defects (quantity of images); ¶ 021, ll 1-14: model determination system according to the present invention, a candidate model extraction module may extract at least two or more candidate models from among a plurality of learning models stored in a learning-model storage on the basis of determination type information; (storage address))
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for product defect analysis response message comprises one or more of: identity, accuracy, and recall of the defect identification model, wherein defect identification model is determined based on product type, the storage address of the product images, and quantity of the product images as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 13, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 7.
Mukundan does not explicitly disclose for a) receiving product defect analysis response message from product manufacturing assisting device, wherein the product defect analysis response message comprises one or more of: product image identity, product defect location, product defect identity, and repair identity, and for b) wherein product defect location, the product defect identity and the repair identity are determined based on the product type, the storage address of the product images, and the quantity of the product images.
However, Tae discloses wherein the receiving the product defect analysis response message comprising:
a) receiving the product defect analysis response message from the product manufacturing assisting device, wherein the product defect analysis response message comprises one or more of: product image identity, product defect location, product defect identity, and repair identity; b) wherein the product defect location, the product defect identity and the repair identity are determined based on the product type, the storage address of the product images, and the quantity of the product images. (Tae ¶ 015, ll 1-6: determination accuracy of a new learning model (response information) may be determined by various factors, wherein the amount of training data, selection of hyper-parameters initially set for learning, selection of a learning model, and the like are important factors in determining the determination accuracy; ¶ 058, ll 1-6: The present disclosure provides an example in which determination type information includes at least one piece of information including: defect type information for the above-described type of defects, product type information about the type of products to be inspected, and part type information about the type of parts to be inspected; ¶ 013, ll 1-8: a format of an image to be used as training data may be different, type of defects may be different depending on the result even with the same defect type, and the result itself revealing a defect may be different, and thus form of the image is different, whereby each learning model should be generated and applied theoretically according to the result of each process and the type of defects (quantity of images); ¶ 021, ll 1-14: model determination system according to the present invention, a candidate model extraction module may extract at least two or more candidate models from among a plurality of learning models stored in a learning-model storage on the basis of determination type information; (storage address))
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for a) receiving product defect analysis response message from product manufacturing assisting device, wherein the product defect analysis response message comprises one or more of: product image identity, product defect location, product defect identity, and repair identity, and for b) wherein product defect location, the product defect identity and the repair identity are determined based on the product type, the storage address of the product images, and the quantity of the product images as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 14, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 1.
Mukundan does not explicitly disclose monitoring one or more of: accuracy, precision, recall, F-score, and speed of the product manufacturing assistance device in processing the product defect analysis tasks.
However, Tae discloses, wherein further comprising: monitoring one or more of: accuracy, precision, recall, F-score, and speed of the product manufacturing assistance device in processing the product defect analysis tasks. (Tae ¶ 036, ll 1-5: provides an effect of increasing determination accuracy of a new learning model by determining an optimal pre-training model in generating the new learning model by using a previously registered learning model as the pre-training model; (selected: accuracy in processing defect analysis tasks))
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for monitoring one or more of: accuracy, precision, recall, F-score, and speed of the product manufacturing assistance device in processing the product defect analysis tasks as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 15, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 1, further comprising:
a) obtaining analysis result data from the plurality of the product defect analysis tasks; (Mukundan ¶ 006, ll 14-19: A measure of defects per unit is then determined based on the product test data retrieved from the database server; The defects per unit is displayed through the display unit: (analysis result data))
b) integrating the analysis result data based on one or more of: the product defect type, a result data format, and a manner in which problems of product defects are resolved. (Mukundan ¶ 013, ll 1-4: FIG. 7 is a screenshot of an exemplary interface of the user computing device of FIG. 1 for selecting a type of product for which the defects per unit is determined; (product type); ¶ 018, ll 1-14: creating a new repository branch containing updates to the package manifests of the repositories, building and testing the new repository branch, merging the new repository branch, then triggering packages with dependency updates to be published (effectively solving the choreography problem))
Regarding Claim 16, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 15, further comprising: sending a product defect alert based on the analysis result data. (Mukundan ¶ 035: the defects per unit of the product is determined; the calculation unit of the user-computing device calculates the defects per unit based on the product test data; ¶ 036: the display unit of the user-computing device displays the defects per unit of the product; the display unit displays the defects per unit of the product in a graph; (notification of defect information upon display); ¶ 054, ll 1-12: if the validation was unsuccessful, the branch merging module invokes the notification module to generate an automated notification (for example email, dashboard notice, or mobile application alert) to the software engineers who made changes in the dependency package as determined by the machine-readable changelog file specified in the present disclosure; (notification, alert))
Regarding Claim 17, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 3.
Mukundan does not explicitly disclose updating the defect identification model.
However, Tae discloses wherein further comprising: updating the defect identification model. (Tae ¶ 110 ll 1-7: The client control module 35 communicates with the vision AI cloud 10 through the client communication module; The client control module controls the overall vision inspection through the deep learning inspection module, and may update a currently registered learning model or register a new learning model through communication with the vision AI cloud)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for updating the defect identification model as taught by Tae. One of ordinary skill in the art would have been motivated to employ the teachings of Tae for the benefits achieved from a system that enables the determination of a level of accuracy for learning models and a determination of an optimum learning model. (Tae ¶ 036; ¶ 031, ll 1-7; ¶ 017, ll 1-7)
Regarding Claim 19, Mukundan-Tae discloses an electronic device comprising: a processor; and a memory, the memory storing computer instructions which, when executed by the processor, implement the method according to claim 1. (Mukundan ¶ 021, ll 1-11: The first memory unit comprises suitable logic, circuitry, interfaces, and/or code that is configured to store the set of instructions, which are executed by the first processor to perform predetermined operation on the user-computing device; the first memory unit is configured to store one or more programs, routines, or scripts that are executed by the first processor in conjunction with the calculation unit)
Regarding Claim 20, Mukundan-Tae discloses a non-transient computer-readable storage medium with computer instructions stored thereon, when the computer instructions are executed by a processor, the method according to claim 1. (Mukundan ¶ 021, ll 1-11: The first memory unit comprises suitable logic, circuitry, interfaces, and/or code that is configured to store the set of instructions, which are executed by the first processor to perform predetermined operation on the user-computing device; the first memory unit is configured to store one or more programs, routines, or scripts that are executed by the first processor in conjunction with the calculation unit)
6. Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Mukundan in view of Tae and further in view of Sawlani et al. (US PGPUB No. 20200226742).
Regarding Claim 4, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 3.
Mukundan does not explicitly disclose a defect identification model comprises any one or more of: a feedforward neural network defect identification model, a convolutional neural network model, a recurrent neural network model, and a generative adversarial network model.
However, Sawlani discloses wherein, the defect identification model comprises any one or more of: a feedforward neural network defect identification model, a convolutional neural network model, a recurrent neural network model, and a generative adversarial network model. (Sawlani ¶ 058, ll 1-11: A “first stage defect classification engine” takes as inputs metrology data such as images of defects, spectra of defects (and their surroundings), and/or wafer maps; A first stage defect classification engine may be implemented by any of various classification algorithms, such as machine learning models; Examples include convolutional neural networks, recurrent neural networks, recurrent convolutional neural networks, generative adversarial networks, autoencoders, and etc.; (selected: a generative adversarial network model))
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for a defect identification model comprises any one or more of: a feedforward neural network defect identification model, a convolutional neural network model, a recurrent neural network model, and a generative adversarial network model as taught by Sawlani. One of ordinary skill in the art would have been motivated to employ the teachings of Sawlani for the flexibility of a system that enables the processing of product defect information utilizing mechanisms such as a generative adversarial network model. (Sawlani ¶ 058, ll 1-11)
7. Claim 18 is rejected under 35 U.S.C. 103 as being unpatentable over Mukundan in view of Tae and further in view of Yuan et al. (US PGPUB No. 20120027288).
Regarding Claim 18, Mukundan-Tae discloses the method for processing product manufacturing messages according to claim 17,
wherein the updating the defect identification model further comprises:
b) distributing a product defect analysis task for online testing of the first defect identification model to obtain a second product defect analysis result; c) in a case where the first defect analysis result and the second defect analysis result meet a predetermined criterion, distributing a product defect analysis task for replacing the second defect identification model in the product manufacturing assisting device with the first defect identification model. (Mukundan ¶ 031, ll 1-8: an input is received through the user-interface displayed on the display unit; the input corresponds to NPI ID of the product for which the defects per unit needs to be calculated; ¶ 044, ll 1-25: Upon receiving the NPI ID, sends a query message to the second database server; The query message corresponds to a query to retrieve the product test data from the database server; sends a retrieval session message to the database server; The retrieval session message corresponds to creation of the data retrieval session between the database server and the user-computing device; ¶ 045, ll 1-9: After termination of the data updating session, the database server sends a retrieval message to the user-computing device; The retrieval message corresponds to transmitting the product test data of the product from the database server to the user-computing device; Upon receiving the product test data from the database server, the calculation unit of the user-computing device calculates and determines defects per unit of the product)
Mukundan does not explicitly disclose for a) offline testing of a first defect identification model.
However, Yuan discloses:
a) distributing a product defect analysis task for offline testing of a first defect identification model to obtain a first product defect analysis result. (Yuan ¶ 023, ll 1-6: relate to defect detection; More specifically, embodiments of the present invention relate to image-based automatic detection of a defective area in an industrial product; ¶ 025, ll 1-6: an offline training stage a model-image database may be created and updated; The model-image database may contain different model images and associated information for each model image)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Mukundan for a) offline testing of a first defect identification model as taught by Yuan. One of ordinary skill in the art would have been motivated to employ the teachings of Yuan for the benefits achieved from a system that enables defect identification processing within an offline environment. (Yuan ¶ 023, ll 1-6; ¶ 025, ll 1-6)
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Kyung H Shin whose telephone number is (571)272-3920. The examiner can normally be reached M - F: 12pm - 8pm.
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, Joon H Hwang can be reached at 571-272-4036. 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.
/KYUNG H SHIN/ 9-23-2026Primary Examiner, Art Unit 2447
Prosecution Insights