DETAILED ACTIONS
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 .
Priority
Acknowledgment is made of applicant’s claim this application being in benefit of foreign priority from Korean Patent Application No. KR10-2023-0053292 filed on April 24, 2023.
Information Disclosure Statement
The information disclosure statement (“IDS”) filed on 01/18/2024 and 01/26/2026 were reviewed and the listed references were noted.
Drawings
The 24-page drawings have been considered and placed on record in the file.
Status of Claims
Claims 1-20 are pending.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-2, 4-6, 8-11, 13-15, and 17-18 are rejected under 35 U.S.C. 103 as being unpatentable over Shin et al., (US 2024/0265519 A1, earliest filing date of the foreign application is 02/07/2023), hereinafter referred to as Shin, in view of Ko et al., "Rigid and non-rigid motion artifact reduction in X-ray CT using attention module" (2021), hereinafter referred to as Ko, in further view of Ge et al., (US 2022/0189005 A1), hereinafter referred to as Ge..
Claim 1
Shin discloses a terminal (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”) comprising:
a display (Shin, [0202], “the electrode state determination unit 307 may process to output a preset notification through the device 100 or an output unit (e.g., a display or speaker) connected to the device 100”); and
a processor (Shin, [0077], “may include one or more communication processors”) is configured to:
generate a reconstructed image based on a plurality of X-ray images of a battery (Shin, [0119], “The image acquisition unit 303 may generate a 3-dimensional (3D) image (or 3-dimensional CT image) corresponding to the specimen 21 by reconstructing a plurality of the acquired tomography images”, CT images is analogous to reconstructed X-ray images),
input the image (Shin, [0021], “The image acquisition unit may generate the 3-dimensional image by integrating a plurality of images, which are obtained by tomography of cross-sections of the electrodes inside the specimen in a specific region of edges of the specimen, where the electrodes are disposed, by means of CT.”, Ge discloses denoising/enhancing the images before using it as an input to the defect detection model) to an artificial neural network-based electrode detection model and generate an electrode detection image based on an output of the artificial neural network-based electrode detection model (Shin, [0133], “ the deep learning-based determination model or the rule-based determination model may be configured to extract electrodes and positions of the electrodes from the tomography images”),
the electrode detection image including a positive electrode of the battery and a negative electrode of the battery in which the positive electrode and the negative electrode are separated from each other (Shin, [0133], “the deep learning-based determination model or the rule-based determination model may be configured to extract the electrodes by distinguishing between the cathode electrode (or cathode layer, hereinafter, cathode) and the anode electrode (or anode layer, hereinafter, anode).”),
perform post-processing on the electrode detection image to generate a post-processing result (Shin, [0139], “the deep learning-based determination model or the rule-based determination model may be configured to extract endpoint positions for each of the electrodes from the tomography images. To describe with reference to FIG. 6, when it is configured to extract electrodes including the electrode 611, the electrode 613, the electrode 621 and the electrode 623 from the first plane tomography image 600, the deep learning-based determination model or the rule-based determination model may be configured to extract endpoint positions of the electrodes, including an electrode endpoint 631, an electrode endpoint 641, an electrode endpoint 643 and an electrode endpoint 633 for each of the extracted electrodes.”), and
detect whether the battery is defective based on the post-processing result (Shin, [0145], “The deep learning-based determination model or the rule-based determination model may be configured to determine whether the calculated gap (value) between two electrode endpoints satisfies a preset endpoint gap reference value (or reference gap). Here, the endpoint reference gap may be set as a value, or may be set as a range (e.g., endpoint gap reference range).”, [0147], “When the calculated gap between two electrode endpoints satisfies a condition of the preset endpoint reference gap (or reference range), the deep learning-based determination model or the rule-based determination model may determine that the corresponding electrodes are in a normal state, and when it does not satisfy the condition, determine that the corresponding electrodes are in an abnormal state”, [0053], “a method for determining the presence or absence of an abnormality in electrodes formed in the battery in a manufacturing process of a secondary battery, for example, misalignment of the electrodes, electrode omission, electrode duplication, and electrode deformation (e.g., twisting or bending, etc.), and detecting a defect of battery, as well as a device therefor will be described.”)
Shin does not explicitly disclose to rotate the reconstructed image by a predetermined angle to generate a tilting image and to input the tilting image to an artificial neural network-based image quality improvement model and generate an output image based on an output of the artificial neural network-based image quality improvement model.
However, Ko teaches to rotate the reconstructed image by a predetermined angle to generate a tilting image (Ko, Section 3, “The 256 ×256 sized CT images were reconstructed using the fil- tered backpropagation (FBP) algorithm. Finally, the images were randomly flipped (left and right) and rotated in the range of [ −10 ◦, 10 ◦].”, CT images is analogous to reconstructed x-ray images, Fig. 2, Zhang discloses that the CT image is a reconstructed X-ray images of a battery) and to input the tilting image to an artificial neural network-based image quality improvement model (Ko, Abstract, “we proposed a new real-time technique for motion artifacts reduction that utilizes a deep residual network with an attention module.”, Section 2, “we compensated for motion artifacts in CT images using a deep neural network (DNN) model.”) and generate an output image based on an output of the artificial neural network-based image quality improvement model (Ko, Fig. 1, “It takes a single motion-corrupted CT image as input, processes it by N AttBlocks, and outputs a motion-reduced CT image”, Section 6, “our motion scenario during training covers the wide range of translation and rotation along the 3-D axis”, Section 2.1, “Our goal was to transform a motion-distorted CT image into an artifact-reduced CT image.”, Ko teaches improving the quality of the image by reducing the artifact created by the rotation of the CT images.”) .
Shin and Ko are both considered to be analogous to the claimed invention because they are in the same field of CT image detection. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the terminal as taught by Shin to incorporate the teachings of Ko to rotate the reconstructed image by a predetermined angle to generate a tilting image and to input the tilting image to an artificial neural network-based image quality improvement model and generate an output image based on an output of the artificial neural network-based image quality improvement model. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been because the network model is robust for more challenging 6-DoF rigid motions that better reflects real patient characteristic, which is in Shin’s case the battery’s characteristics and the model also works for different CT geometries as proven by using it on CBCT teeth dataset (Ko, Section 5.3).
The combination of Shin in view of Ko does not explicitly disclose to input the image to an artificial neural network-based image quality improvement model and generate an output image based on an output of the artificial neural network-based image quality improvement model, input the output image to an artificial neural network-based electrode detection model, and detect whether the battery is defective based on the post-processing result.
However, Ge teaches to input the image (Ge, [0018], “CT imaging techniques can be used to inspect industrial machines (e.g., batteries) and identify characteristics in the industrial machines (e.g., detect defects in the batteries).”, Ko teaches rotating the image to a predefined angle) to an artificial neural network-based image quality improvement model (Ge, [0032], “the denoising/upscaling algorithm can include AI models (e.g., deep learning networks) trained on a dataset of noisy images and configured to identify and/or filter out noise characteristics from the image”) and generate an output image based on an output of the artificial neural network-based image quality improvement model (Ge, [0020], “capturing high resolution image may need longer scanning time (which can lead to higher scanning costs) while a fast scan may result in poor image quality. In some implementations, the defect detection system can apply AI based image denoising and upscaling filters on the scanned images. This can allow for defect detection on images that may not be prepared for AI based defect detection using traditional image enhancing techniques”), input the output image to an artificial neural network-based electrode detection model (Ge, [0019], “the defect detection system can automatically select an AI model (e.g., a deep learning model) from a plurality of Al models based on the type of inspection to be performed (e.g., determination of the presence of a defect, identification of a defect type, determination of a defect characteristic, etc.)”, [0020], “the defect detection system can apply AI based image denoising and upscaling filters on the scanned images. This can allow for defect detection on images that may not be prepared for AI based defect detection using traditional image enhancing techniques”, Shin discloses extracting the positive and negative electrode first before determining the defect on the battery) detect whether the battery is defective based on the post-processing result (Ge, [0019], the defect detection system can automatically select an AI model (e.g., a deep learning model) from a plurality of Al models based on the type of inspection to be performed (e.g., determination of the presence of a defect, identification of a defect type, determination of a defect characteristic, etc.). The plurality of AI models can be pre-trained and can be included in an AI model library. The various AI models can be trained to detect defect in one or more batteries e.g., batteries of given battery types, battery models, etc.), various scanning environments, defect modes and original equipment manufacturers (OEMs) of batteries.”).
Shin, Ko, and Ge are all considered to be analogous to the claimed invention because they are in the same field of CT image detection. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the terminal as taught by Shin to incorporate the teachings of Ge to input the image to an artificial neural network-based image quality improvement model and generate an output image based on an output of the artificial neural network-based image quality improvement model, input the output image to an artificial neural network-based electrode detection model, and detect whether the battery is defective based on the post-processing result. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to reduce inspection costs while providing comprehensive inspection that can lead to competitive advantage in the growing CT inspection market because it allows for defect detection on images that may not be prepared for AI based defect detection using traditional image enhancing techniques (Ge, [0020]).
Claim 2
The combination of Shin in view of Ko in further view of Ge discloses a terminal of claim 1 (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”), wherein the battery is a secondary battery or a rechargeable battery (Shin, [0053], “a method for determining the presence or absence of an abnormality in electrodes formed in the battery in a manufacturing process of a secondary battery”).
Claim 4
The combination of Shin in view of Ko in further view of Ge discloses a terminal of claim 1 (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”), wherein the processor (Shin, [0077], “may include one or more communication processors”) is further configured to perform a skeletonization operation on the positive electrode and the negative electrode included in the electrode detection image (Shin, Fig. 6, [0139], “the deep learning-based determination model or the rule-based determination model may be configured to extract endpoint positions of the electrodes, including an electrode endpoint 631, an electrode endpoint 641, an electrode endpoint 643 and an electrode endpoint 633 for each of the extracted electrodes.”).
Claim 5
The combination of Shin in view of Ko in further view of Ge discloses a terminal of claim 4 (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”), wherein the skeletonization operation includes extracting pixels located on a center line of each of the positive electrode and the negative electrode image (Shin, Fig. 6, [0139], “the deep learning-based determination model or the rule-based determination model may be configured to extract endpoint positions of the electrodes, including an electrode endpoint 631, an electrode endpoint 641, an electrode endpoint 643 and an electrode endpoint 633 for each of the extracted electrodes.”), and wherein the processor (Shin, [0077], “may include one or more communication processors”) is further configured to: calculate a difference between a number of pixels of the positive electrode and a number of pixels of the negative electrode (Shin, [0133], “the deep learning-based determination model or the rule-based determination model may be configured to extract the electrodes by distinguishing between the cathode electrode (or cathode layer, hereinafter, cathode) and the anode electrode (or anode layer, hereinafter, anode).”, [0174], “identify the cathode and anode of the extracted electrodes, and detect a state where an electrode (e.g., cathode) having a different polarity is omitted between an electrode 811 (e.g., anode) and an electrode 813 (e.g., anode) based on the identified cathodes and anodes”, by identifying the cathode and anode it also identifies the difference between the number of each), calculate a length difference between the positive electrode and the negative electrode based on the length difference (Shin, [0140], “the deep learning-based determination model or the rule-based determination model may be configured to calculate a gap between adjacent electrode endpoints based on the extracted electrode endpoints. In this case, the deep learning-based determination model or the rule-based determination model may be configured to calculate a gap between adjacent electrode endpoints in each of the first plane tomography image and the second plane tomography image”, [0153], “two electrode endpoints adjacent to each other are described as an example of electrodes having different polarities”), in response to the length difference being less than a predetermined value, determine that the battery is non-defective (Shin, [0149], “if the calculated gap between two electrode endpoints is smaller (or not more) than the endpoint reference gap, the deep learning-based determination model or the rule-based determination model may be configured to determine that the corresponding electrodes are in a normal state.”), and in response to the length difference exceeding the predetermined value, determine that the battery is defective (Shin, [0149], “if the calculated gap between two electrode endpoints is smaller (or not more) than the endpoint reference gap, the deep learning-based determination model or the rule-based determination model may be configured to determine that the corresponding electrodes are in a normal state.” [0147], “When the calculated gap between two electrode endpoints satisfies a condition of the preset endpoint reference gap (or reference range), the deep learning-based determination model or the rule-based determination model may determine that the corresponding electrodes are in a normal state, and when it does not satisfy the condition, determine that the corresponding electrodes are in an abnormal state.”, the condition for it to be non-defective is for the gap to be less than the reference gap otherwise it is deemed abnormal or defective).
Claim 6
The combination of Shin in view of Ko in further view of Ge discloses a terminal of claim 1 (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”), wherein the processor (Shin, [0077], “may include one or more communication processors”) is further configured to display a detection result of whether or not the battery is defective on the display (Shin, [0200], “the electrode state determination unit 307 may display the electrode that has been determined in the abnormal state on at least one tomography image by the bounding box, and may label it with a category of misalignment, electrode omission, electrode duplication or electrode deformation as a cause that has determined the abnormality in the corresponding electrodes”).
Claim 8
The combination of Shin in view of Ko in further view of Ge discloses a terminal of claim 1 (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”), further comprising: a memory (Ko, [0074], “storage unit 120 may include a volatile memory, a non-volatile memory, and/or a computer-readable recording medium as known in the art”) configured to store the artificial neural network-based image quality improvement model (Ko, Abstract, “we proposed a new real-time technique for motion artifacts reduction that utilizes a deep residual network with an attention module.”, Section 2, “we compensated for motion artifacts in CT images using a deep neural network (DNN) model.”) , and store a training data set used for supervised learning of the artificial neural network-based image quality improvement model (Ko, Section 6, “we use strong supervisions earned by a carefully collected paired training dataset, where the target signal is well specified (e.g. head CT) and the target variation is also well constrained (e.g. patient motions)”), the training data set including the tilting image for training and a labeling image labeled based on the tilting image (Ko, Abstract, “Each dataset provided a set of motion-corrupted CT images and their ground-truth CT image pairs”, Section 6, “our motion scenario during training covers the wide range of translation and rotation along the 3-D axis”.
The proposed combination as well as the motivation for combining the Shin, Ko, and Ge references presented in the rejection of Claim 1, apply to Claim 8 and are incorporated herein by reference. Thus, the terminal recited in Claim 8 is met by Shin, Ko, and Ge.
Claim 9
The combination of Shin in view of Ko in further view of Ge discloses a terminal of claim 1 (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”), further comprising: a communication interface (Shin, [0047], “it will be understood that when a component is described as being “connected” or “combined” by communication to another component, that component may be connected or combined by wireless or wired communication to the another component, but it may be “connected” or “combined” to the another component intervening another component may be present.”) configured to receive the plurality of X-ray images (Shin, [0020], “an image acquisition unit configured to acquire a 3-dimensional image by imaging a specimen including one or more battery cell type electrodes”) from an X-ray imaging device (Shin, [0055], “an electrode inspection device for performing CT scan of the specimen using X-rays”, [0092], “The source unit 201 of the electrode inspection device may include a device for generating and emitting X-rays, and the detection unit 203 may include a device for detecting the X-rays emitted from the source unit 201. In this case, the electrode inspection device may be provided with two or more source units 201 or detection units 203”)
Claims 10-11, 13-15, and 17-18 are rejected for similar reasons as those described in claims 1-2, 4-6, and 8-9. The additional elements in Claims 10-11, 13-15, and 17-18 (the combination of Shin in view of Ko in further view of Ge) discloses includes: a method of controlling a terminal for defect detection (Shin, Fig. 11). The proposed combination as well as the motivation for combining the Shin, Ko, and Ge references presented in the rejection of Claim 1, apply to Claims 10-11, 13-15, and 17-18 and are incorporated herein by reference. Thus, the method recited in Claims 10-11, 13-15, and 17-18 is met by Shin, Ko, and Ge.
Claims 3, 7, 12, and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Shin in view of Ko in view of Ge in further view of Muller et al., " Deep learning-based segmentation of lithium-ion battery microstructures enhanced by artificially generated electrodes" (2021), hereinafter referred to as Muller.
Claim 3
The combination of Shin in view of Ko in further view of Ge discloses a terminal of claim 1 (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”), wherein the artificial neural network-based image quality improvement model (Ko, Abstract, “we proposed a new real-time technique for motion artifacts reduction that utilizes a deep residual network with an attention module.”, Section 2, “we compensated for motion artifacts in CT images using a deep neural network (DNN) model.”) is learned through a supervised learning (Ko, Section 6, “we use strong supervisions earned by a carefully collected paired training dataset, where the target signal is well specified (e.g. head CT) and the target variation is also well constrained (e.g. patient motions)”).
The combination of Shin in view of Ko in further view of Ge does not explicitly disclose wherein the artificial neural network-based electrode detection model is learned through a supervised learning.
However, Muller teaches wherein the artificial neural network-based electrode detection model is learned through a supervised learning (Muller, page 2, “we show how supervised, deep learning can help address the challenges associated with semantic segmentation of high-resolution, volumetric image data of LIB electrodes.”, Abstract, “we demonstrate a methodology for using deep learning tools to achieve reliable segmentations of volumetric images of electrodes on which standard segmentation approaches fail due to insufficient contrast. We implement the 3D U-Net architecture for segmentation, and, to overcome the limitations of training data obtained experimentally through imaging, we show how synthetic learning data, consisting of realistic artificial electrode structures and their tomographic reconstructions, can be generated and used to enhance network performance. We apply our method to segment x-ray tomographic microscopy images of graphite-silicon composite electrodes and show it is accurate across standard metrics. “).
Shin, Ko, Ge, and Muller are all considered to be analogous to the claimed invention because they are in the same field of CT image detection. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the terminal as taught by Shin to incorporate the teachings of Muller wherein the artificial neural network-based electrode detection model is learned through a supervised learning. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to help address the challenges associated with semantic segmentation of high-resolution, volumetric image data of LIB electrodes (Muller, page 2).
Claim 7
The combination of Shin in view of Ko in further view of Ge discloses a terminal of claim 1 (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”).
The combination of Shin in view of Ko in further view of Ge does not explicitly disclose a memory configured to store the artificial neural network-based electrode detection model, and store a training data set used for supervised learning of the artificial neural network-based electrode detection model, the training data set including the output image for training and a labeling image labeled based on the output image.
However, Muller teaches a memory (Muller, page 5, “Nvidia TITAN Xp GPU”) configured to store the artificial neural network-based electrode detection model (Muller, page 2, “we show how supervised, deep learning can help address the challenges associated with semantic segmentation of high-resolution, volumetric image data of LIB electrodes.”, Abstract, “we demonstrate a methodology for using deep learning tools to achieve reliable segmentations of volumetric images of electrodes on which standard segmentation approaches fail due to insufficient contrast. We implement the 3D U-Net architecture for segmentation, and, to overcome the limitations of training data obtained experimentally through imaging, we show how synthetic learning data, consisting of realistic artificial electrode structures and their tomographic reconstructions, can be generated and used to enhance network performance. We apply our method to segment x-ray tomographic microscopy images of graphite-silicon composite electrodes and show it is accurate across standard metrics. “), and store a training data set used for supervised learning of the artificial neural network-based electrode detection model, the training data set including the output image for training and a labeling image labeled based on the output image (Muller, page 2, “we propose and demonstrate the benefits of using hybrid learning datasets (Fig. 1), where computationally generated synthetic datasets augment a limited number of real datasets, acquired using multimodal imaging techniques.”, page 5, “To quantify the segmentation quality, 400 sequential slices of the XTM dataset of one of the three pristine samples are manually segmented with the help of the Dragonfly software. This manually segmented volume serves as the “ground truth” and is otherwise not used for training. We evaluate segmentation quality according to nineteen standard metrics based on similarity and distance criteria40, where the learning-based segmentation of one pristine sample are compared to the corresponding “ground truth” (Table S2)”).
Shin, Ko, Ge, and Muller are all considered to be analogous to the claimed invention because they are in the same field of CT image detection. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the terminal as taught by Shin to incorporate the teachings of Muller of a memory configured to store the artificial neural network-based electrode detection model, and store a training data set used for supervised learning of the artificial neural network-based electrode detection model, the training data set including the output image for training and a labeling image labeled based on the output image. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to help address the challenges associated with semantic segmentation of high-resolution, volumetric image data of LIB electrodes (Muller, page 2).
Claims 12 and 16 are rejected for similar reasons as those described in claims 3 and 7. The additional elements in Claims 12 and 16 (the combination of Shin in view of Ko in further view of Ge in view of Muller) discloses includes: a method of controlling a terminal for defect detection (Shin, Fig. 11). The proposed combination as well as the motivation for combining the Shin, Ko, Ge, and Muller references presented in the rejection of Claim 1, apply to Claims 12 and 16 are incorporated herein by reference. Thus, the method recited in Claims 12 and 16 is met by Shin, Ko, Ge, and Muller.
Claims 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Shin et al., (US 2024/0265519 A1, earliest filing date of the foreign application is 02/07/2023), hereinafter referred to as Shin, in view of Rahman et al., "A Deep Learning Approach for Liver and Tumor Segmentation in CT Images Using Res UNet" (July 2022), hereinafter referred to as Rahman.
Claim 19
Shin discloses a terminal (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”) comprising:
a display (Shin, [0202], “the electrode state determination unit 307 may process to output a preset notification through the device 100 or an output unit (e.g., a display or speaker) connected to the device 100”); and
a processor (Shin, [0077], “may include one or more communication processors”) is configured to:
generate a reconstructed image based on a plurality of X-ray images of a battery (Shin, [0119], “The image acquisition unit 303 may generate a 3-dimensional (3D) image (or 3-dimensional CT image) corresponding to the specimen 21 by reconstructing a plurality of the acquired tomography images”, CT images is analogous to reconstructed X-ray images),
generate an electrode detection image (Shin, [0021], “The image acquisition unit may generate the 3-dimensional image by integrating a plurality of images, which are obtained by tomography of cross-sections of the electrodes inside the specimen in a specific region of edges of the specimen, where the electrodes are disposed, by means of CT.”) based on the image (Shin, [0133], “ the deep learning-based determination model or the rule-based determination model may be configured to extract electrodes and positions of the electrodes from the tomography images”), the electrode detection image including a positive electrode of the battery and a negative electrode of the battery in which the positive electrode and the negative electrode are separated from each other (Shin, [0133], “the deep learning-based determination model or the rule-based determination model may be configured to extract the electrodes by distinguishing between the cathode electrode (or cathode layer, hereinafter, cathode) and the anode electrode (or anode layer, hereinafter, anode).”), and
detect whether the battery is defective based on the electrode detection image (Shin, [0145], “The deep learning-based determination model or the rule-based determination model may be configured to determine whether the calculated gap (value) between two electrode endpoints satisfies a preset endpoint gap reference value (or reference gap). Here, the endpoint reference gap may be set as a value, or may be set as a range (e.g., endpoint gap reference range).”, [0147], “When the calculated gap between two electrode endpoints satisfies a condition of the preset endpoint reference gap (or reference range), the deep learning-based determination model or the rule-based determination model may determine that the corresponding electrodes are in a normal state, and when it does not satisfy the condition, determine that the corresponding electrodes are in an abnormal state”, [0053], “a method for determining the presence or absence of an abnormality in electrodes formed in the battery in a manufacturing process of a secondary battery, for example, misalignment of the electrodes, electrode omission, electrode duplication, and electrode deformation (e.g., twisting or bending, etc.), and detecting a defect of battery, as well as a device therefor will be described.”)
Shin does not explicitly disclose to rotate the reconstructed image by a predetermined angle to generate a tilting image, generate an electrode detection image based on the tilting image.
However, Rahman teaches to rotate the reconstructed image by a predetermined angle to generate a tilting image (Rahman, Figure 1, the CT image input is rotated prior to the region of interest segmentation, the CT image is analogous to the reconstructed image, Fig. 10, rotation of the CT images for 90 degrees), generate an electrode detection image based on the tilting image (Rahman, Fig. 1, teaches detecting the tumor and the liver, Shin teaches that the CT images is of a battery and the detecting the electrodes).
Shin and Rahman are both considered to be analogous to the claimed invention because they are in the same field of CT image detection. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the terminal as taught by Shin to incorporate the teachings of Ko to rotate the reconstructed image by a predetermined angle to generate a tilting image, generate an electrode detection image based on the tilting image. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to enhance the quality of the raw input image (Rahman, Section 3.1).
Claim 20
The combination of Shin in view of Rahman discloses the terminal of claim 19 (Shin, [0082, “the device 100 or the user device may include at least some of functions of all information and communication devices including a mobile communication terminal, a multimedia terminal, a wired terminal, a fixed terminal, an internet protocol (IP) terminal and the like.”), wherein the processor (Shin, [0077], “may include one or more communication processors”) is further configured to: perform a skeletonization operation on the positive electrode and the negative electrode included in the electrode detection image by extracting pixels located on a center line of each of the positive electrode and the negative electrode (Shin, Fig. 6, [0139], “the deep learning-based determination model or the rule-based determination model may be configured to extract endpoint positions of the electrodes, including an electrode endpoint 631, an electrode endpoint 641, an electrode endpoint 643 and an electrode endpoint 633 for each of the extracted electrodes.”), calculate a difference between a number of pixels of the positive electrode and a number of pixels of the negative electrode (Shin, [0133], “the deep learning-based determination model or the rule-based determination model may be configured to extract the electrodes by distinguishing between the cathode electrode (or cathode layer, hereinafter, cathode) and the anode electrode (or anode layer, hereinafter, anode).”, [0174], “identify the cathode and anode of the extracted electrodes, and detect a state where an electrode (e.g., cathode) having a different polarity is omitted between an electrode 811 (e.g., anode) and an electrode 813 (e.g., anode) based on the identified cathodes and anodes”, by identifying the cathode and anode it also identifies the difference between the number of each), calculate a length difference between the positive electrode and the negative electrode based on the length difference (Shin, [0140], “the deep learning-based determination model or the rule-based determination model may be configured to calculate a gap between adjacent electrode endpoints based on the extracted electrode endpoints. In this case, the deep learning-based determination model or the rule-based determination model may be configured to calculate a gap between adjacent electrode endpoints in each of the first plane tomography image and the second plane tomography image”, [0153], “two electrode endpoints adjacent to each other are described as an example of electrodes having different polarities”), in response to the length difference being less than a predetermined value, determine that the battery is non-defective (Shin, [0149], “if the calculated gap between two electrode endpoints is smaller (or not more) than the endpoint reference gap, the deep learning-based determination model or the rule-based determination model may be configured to determine that the corresponding electrodes are in a normal state.”), and in response to the length difference exceeding the predetermined value, determine that the battery is defective (Shin, [0149], “if the calculated gap between two electrode endpoints is smaller (or not more) than the endpoint reference gap, the deep learning-based determination model or the rule-based determination model may be configured to determine that the corresponding electrodes are in a normal state.” [0147], “When the calculated gap between two electrode endpoints satisfies a condition of the preset endpoint reference gap (or reference range), the deep learning-based determination model or the rule-based determination model may determine that the corresponding electrodes are in a normal state, and when it does not satisfy the condition, determine that the corresponding electrodes are in an abnormal state.”, the condition for it to be non-defective is for the gap to be less than the reference gap otherwise it is deemed abnormal or defective).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DENISE G ALFONSO whose telephone number is (571)272-1360. The examiner can normally be reached Monday - Friday 7:30 - 5:30.
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, Amandeep Saini can be reached at (571)272-3382. 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.
/DENISE G ALFONSO/Examiner, Art Unit 2662
/AMANDEEP SAINI/Supervisory Patent Examiner, Art Unit 2662