Battery Electrode Inspection Device and Method

Abstract

The battery electrode inspection device according to an embodiment of the present invention extracts an inspection target image including a region suspected of a defect based on an electrode image acquired from a camera, and sets and applies different types of learning models for determining the presence or absence of defects in the electrode corresponding to the inspection target image according to the amount of learning data, thereby shortening the application time due to the initial process application and the process environment change. By re-learning the learning model based on the updated learning data, highly reliable result data with improved accuracy can be obtained.

Description

Technical Field

[0001] This application claims the benefit of the filing dates of Korean Patent Application No. 10-2022-0109765, filed with the Korean Intellectual Property Office on August 31, 2022, and Korean Patent Application No. 10-2023-0103916, filed with the Korean Intellectual Property Office on August 9, 2023, and all of the contents disclosed in the documents of the Korean patent applications are incorporated herein.

[0002] The present invention relates to a battery electrode inspection apparatus and method, and more particularly, to a battery electrode inspection apparatus and method for determining the presence or absence of defects in an electrode due to surface defects from an electrode image using a pre-trained learning model.

Background Art

[0003] Due to the depletion of fossil fuels, the price of energy sources has risen, and the interest in environmental pollution has increased. As an environmentally friendly alternative energy source, the demand for secondary batteries is rapidly increasing.

[0004] Among secondary batteries, lithium batteries are applied to many industrial fields such as mobile application devices, automobiles, robots, and energy storage devices as a countermeasure against recent environmental regulations and high crude oil prices.

[0005] Such lithium batteries are generally classified into cylindrical, prismatic, or pouch types according to the shape of the exterior material in which the electrode assembly is housed.

[0006] Among these, cylindrical batteries are provided in the structure of a battery pack (Cell To Pack, CTP) composed of a plurality of battery cells. In other words, a cylindrical battery is provided in a form in which an electrode formed from a separator between a positive electrode and a negative electrode is wound and inserted into the inside of a battery can.

[0007] In the electrode process of a cylindrical battery, a battery electrode inspection apparatus is used to detect defects on the electrode surface due to foreign matter, scratches, etc.

[0008] Generally, a battery electrode inspection device analyzes a camera installed in process equipment and electrode images obtained from the camera to determine the presence or absence of defects due to defects on the electrode surface. As a result, the battery electrode inspection device can affect the result data due to on-site conditions or environments such as equipment shaking and process condition changes.

[0009] Therefore, a conventional battery electrode inspection device applied a program logic designed based on defective images obtained from a camera installed on-site during the initial application of the process.

[0010] Therefore, the conventional battery electrode inspection device had the disadvantage of taking a long time during the initial application of the process.

[0011] In addition, the conventional battery electrode inspection device had the disadvantage that when changes occurred in the on-site conditions or environment, the program logic had to be continuously redesigned by an engineer.

Summary of the Invention

Problems to be Solved by the Invention

[0012] An object of the present invention for solving the above problems is to provide a battery electrode inspection system.

[0013] An object of the present invention for solving the above problems is to provide a battery electrode inspection device.

[0014] Another object of the present invention for solving the above problems is to provide a battery electrode inspection method.

Means for Solving the Problems

[0015] An artificial intelligence model-based battery electrode inspection apparatus according to an embodiment of the present invention for achieving the above object includes a memory and a processor that executes at least one instruction stored in the memory. The at least one instruction includes an instruction to obtain an inspection target image by extracting a region suspected of being defective from an electrode image in which at least one electrode surface is photographed, and an instruction to input the inspection target image into a learned learning model and output result data regarding the presence or absence of defects in the battery electrode.

[0016] Here, the instruction to output the result data may include an instruction to determine any one of a plurality of learning models according to at least one learning data amount.

[0017] At this time, the instruction to determine the learning model may include an instruction to output the result data using a first learning model based on machine learning that has already been learned when the learning data amount is less than a preset reference value.

[0018] On the other hand, the instruction to determine the learning model may include an instruction to output the result data using a second learning model based on deep learning that has already been learned when the learning data amount is greater than or equal to the preset reference value.

[0019] On the other hand, the instruction to output the result data using the first learning model may include an instruction to extract at least one image feature value from the inspection target image, and an instruction to input at least one of the image feature values into the first learning model and obtain the result data.

[0020] At this time, the instruction to extract the image feature value may include an instruction to extract the image feature value from the inspection target image using a rule-based learning algorithm.

[0021] Here, the image feature value can include at least one data among the height of the pixel (Height), the width of the pixel (Width), the maximum pixel value (Peak White), the minimum pixel value (Peak Dark), the aspect ratio, and the circularity ratio extracted from the inspection target image.

[0022] In addition, the first learning model may be a learning model based on Random Forest.

[0023] On the other hand, the instruction to output the result data using the second learning model can include an instruction to input at least one inspection target image into the second learning model to obtain the result data.

[0024] At this time, the second learning model may be a learning model based on a Convolutional Neural Network (CNN).

[0025] In addition, the at least one instruction can further include an instruction to re-learn the already learned learning model by utilizing the result data as learning data.

[0026] A battery electrode inspection method using an artificial intelligence model-based battery electrode inspection device according to another embodiment of the present invention for achieving the above object includes a step of obtaining an inspection target image by extracting a region suspected of being defective from an electrode image in which at least one electrode surface is photographed, and a step of inputting the inspection target image into an already learned learning model and outputting result data regarding the presence or absence of defects in the battery electrode.

[0027] Here, the step of outputting the result data can include a step of determining any one of a plurality of learning models according to at least one learning data amount.

[0028] At this time, the step of determining the learning model may include the step of outputting the result data using a first machine learning-based learning model that has already been learned when the amount of the learning data is less than a preset reference value.

[0029] On the other hand, the step of determining the learning model may include the step of outputting the result data using a second deep learning-based learning model that has already been learned when the amount of the learning data is greater than or equal to the preset reference value.

[0030] On the other hand, the step of outputting the result data using the first learning model may include an instruction to extract at least one image feature value from the image to be inspected, and a step of inputting at least one of the image feature values into the first learning model to obtain the result data.

[0031] At this time, the step of extracting the image feature value may include the step of extracting the image feature value from the image to be inspected using a rule-based learning algorithm.

[0032] Here, the image feature value may include at least one data among the height of the pixel, the width of the pixel, the peak white value of the pixel, the peak dark value of the pixel, the aspect ratio, and the circularity ratio extracted from the image to be inspected.

[0033] Also, the first learning model may be a random forest-based learning model.

[0034] On the other hand, the step of outputting the result data using the second learning model may include a step of inputting at least one image to be inspected into the second learning model to obtain the result data.

[0035] At this time, the second learning model may be a learning model based on a Convolutional Neural Network (CNN).

[0036] In addition, the battery electrode inspection method may further include a step of re-learning the already learned learning model by using the result data as learning data.

[0037] An artificial intelligence model-based battery electrode inspection system according to another embodiment of the present invention for achieving the above object includes a camera that captures at least one electrode surface to generate an electrode image, and an inspection target image obtained by extracting a region suspected of being defective from the electrode image, and inputs the inspection target image into an already learned learning model, and includes a battery electrode inspection device that outputs result data regarding the presence or absence of defects in the battery electrode.

Advantages of the Invention

[0038] The battery electrode inspection device, method, and system according to the embodiment of the present invention as described above set different types of learning models according to the amount of learning data to inspect the presence or absence of defects in the inspection target image, and re-learn the learning model based on the learning data in which the result data is reflected, so that the inspection reliability can be improved.

Brief Description of the Drawings

[0039]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Embodiments for Carrying Out the Invention

[0040] Since the present invention can be subjected to various modifications and can have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, but should be understood to include all modifications, equivalents, or alternatives included in the spirit and technical scope of the present invention. Similar reference numerals are used for similar components while explaining each drawing.

[0041] Terms such as first, second, A, B, etc. can be used to describe various components, but the above components should not be limited by the above terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the rights of the present invention, the first component can be named the second component, and similarly, the second component can also be named the first component. The term "and / or" includes a combination of a plurality of related items described or one of a plurality of related items described.

[0042] When a component is referred to as being "coupled" or "connected" to another component, it should be understood that it may be directly coupled or connected to the other component, but there may also be other components in between. In contrast, when a component is referred to as being "directly coupled" or "directly connected" to another component, it should be understood that there are no other components in between.

[0043] The terms used in this application are merely used to describe specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, terms such as "including" or "having" are intended to specify the presence of the features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, and do not preclude the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0044] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the technical field to which the present invention belongs. Terms defined as in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted as an ideal or overly formal meaning unless clearly defined in this application.

[0045] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0046] FIG. 1 is a conceptual diagram of the detection logic application process of a general battery electrode inspection device.

[0047] Referring to FIG. 1, the battery electrode inspection device can detect an electrode with a defect on the electrode surface and determine whether there is a defect.

[0048] A general battery electrode inspection device uses pre-designed detection logic to determine the defect type and presence or absence of defects in at least one electrode image obtained from a camera installed on-site. Here, the detection logic may be logic pre-designed by an engineer using the image features of at least one electrode image for which the defect type has been confirmed.

[0049] FIG. 2 is an operational conceptual diagram of a general battery electrode inspection device when an error occurs.

[0050] Referring to FIG. 2, the detection logic of a general battery electrode inspection device is designed based on a defective image of an electrode obtained from a camera installed on-site at the initial application of the process. As a result, when at least one of the on-site environmental factors or process conditions changes, a general battery electrode inspection device has the drawback that detection errors occur due to distortion or deformation of the electrode image, resulting in a decrease in defect detection accuracy. Therefore, a general electrode detection device has the drawback that the detection logic must be continuously corrected by an engineer.

[0051] Also, when correcting the detection logic to prevent this, in order to consider the above process factors, defective images generated from the camera installed on-site must be obtained to correct the detection logic, which has the drawback of taking a long time.

[0052] And a general battery electrode inspection device has the drawback that it is difficult to determine the classification accuracy for at least one defective image for which the inspection was completed before the correction when the detection logic is corrected. Here, the defective image may be an image determined to be defective due to the occurrence of a defect as a result of the inspection of the detection logic.

[0053] Therefore, in the present invention, an electrode detection device and method will be described that use a pre-trained learning model to extract an inspection target image in which surface defects of an electrode are suspected from an electrode image, and based on this, determine the presence or absence of defects in the corresponding electrode.

[0054] FIG. 3 is a block diagram of a battery electrode inspection system according to an embodiment of the present invention.

[0055] Referring to FIG. 3, the battery electrode inspection system can detect an electrode with a defect on its surface during the electrode process and determine whether the electrode is defective.

[0056] More specifically, it can include a camera 1000 and a battery electrode inspection device 5000.

[0057] The camera 1000 can be installed at the site where the electrode process is performed. For example, the camera 1000 can be fixedly installed on a facility where at least one electrode is sequentially moved. Thereby, the camera 1000 can obtain at least one electrode image by individually photographing the at least one electrode.

[0058] The battery electrode inspection device 5000 can be linked with at least one camera 1000 to obtain at least one electrode image from the camera 1000. However, the battery electrode inspection device 5000 is not limited thereto, and at least one electrode image can be obtained through various routes.

[0059] After that, the battery electrode inspection device 5000 can determine whether the battery electrode is defective based on at least one electrode image by using a learned learning model.

[0060] Hereinafter, the battery electrode inspection device 5000 will be described in more detail.

[0061] FIG. 4 is a block diagram of a battery electrode inspection device according to an embodiment of the present invention.

[0062] Referring to FIG. 4, as described above, the battery electrode inspection apparatus 5000 according to an embodiment of the present invention can extract an image of an inspection target area suspected of having a defect from at least one electrode image, and based on this, can determine whether there is a defect in the electrode by using a pre-trained learning model. Here, the defect may include foreign matter, scratches, cracks, etc.

[0063] Specifically described by component, the battery electrode inspection apparatus 5000 may include a memory 100, a processor 200, a transceiver 300, an input interface device 400, an output interface device 500, and a storage device 600.

[0064] According to an embodiment, each component 100, 200, 300, 400, 500, 600 included in the battery electrode inspection apparatus 5000 can be connected by a bus 700 to communicate with each other.

[0065] Among the above components 100, 200, 300, 400, 500, 600, the memory 100 and the storage device 600 can be composed of at least one of a volatile storage medium and a non-volatile storage medium. For example, the memory 100 and the storage device 600 can be composed of at least one of a read only memory (ROM) and a random access memory (RAM).

[0066] Among them, the memory 100 can include at least one instruction executed by the processor 200.

[0067] According to an embodiment, at least one instruction can include an instruction to obtain an inspection target image that extracts an area suspected of having a defect from an electrode image in which at least one electrode surface is photographed, and an instruction to input the inspection target image into a pre-trained learning model to output result data regarding whether there is a defect in the battery electrode.

[0068] Here, the instruction to output the above result data can include an instruction to determine any one of a plurality of learning models according to at least one amount of learning data.

[0069] At this time, the instruction to determine the above learning model can include an instruction to output the above result data using a first machine learning-based learning model that has already been learned when the above amount of learning data is less than a preset reference value.

[0070] On the other hand, the instruction to determine the above learning model can include an instruction to output the above result data using a second deep learning-based learning model that has already been learned when the above amount of learning data is greater than or equal to the preset reference value.

[0071] On the other hand, the instruction to output the above result data using the above first learning model can include an instruction to extract at least one image feature value from the above image to be inspected, and an instruction to input at least one of the above image feature values into the above first learning model to obtain the above result data.

[0072] At this time, the instruction to extract the above image feature value can include an instruction to extract the above image feature value from the above image to be inspected using a rule-based learning algorithm.

[0073] Here, the above image feature value can include at least one data among the height of the pixel, the width of the pixel, the peak white value of the pixel, the peak dark value of the pixel, the aspect ratio, and the circularity ratio extracted from the above image to be inspected.

[0074] Also, the above first learning model may be a random forest-based learning model.

[0075] On the other hand, the instruction to output the result data using the second learning model may include an instruction to input at least one inspection target image into the second learning model to obtain the result data.

[0076] At this time, the second learning model may be a learning model based on a Convolutional Neural Network (CNN).

[0077] In addition, the at least one instruction may further include an instruction to reuse the result data as learning data to relearn the already learned learning model.

[0078] On the other hand, the processor 200 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor in which the method according to the embodiment of the present invention is performed.

[0079] As described above, the processor 200 can execute at least one program command stored in the memory 100.

[0080] FIG. 5 is a flowchart of a battery electrode inspection method using the battery electrode inspection apparatus according to an embodiment of the present invention.

[0081] Referring to FIG. 5, the battery electrode inspection apparatus 5000 can acquire at least one electrode image (S1000).

[0082] Subsequently, the battery electrode inspection apparatus 5000 can extract an inspection target image from at least one electrode image (S3000). Here, the electrode image may be an image in which the surface of the electrode is photographed by at least one camera 1000 attached to the electrode process equipment.

[0083] Alternatively, the inspection target image may be an image obtained by individually extracting at least one region in which a defect has occurred in at least one electrode image. Here, the defects may include, for example, types such as scratches, foreign substances, and deterioration, but are not limited to what is disclosed.

[0084] Subsequently, the battery electrode inspection apparatus 5000 can determine whether there is a defect in the electrode based on the inspection target image using a previously learned learning model (S5000).

[0085] Subsequently, the battery electrode inspection apparatus 5000 can re-learn the learning model based on the updated at least one learning data (S7000).

[0086] FIG. 6 is an image for explaining the step of extracting an inspection target image in the battery electrode inspection method according to an embodiment of the present invention.

[0087] Referring to FIG. 6, the inspection target image can be obtained using a rule-based algorithm. In other words, the battery electrode inspection apparatus 5000 can input the electrode image and a predefined conditional instruction into the rule-based algorithm to output the inspection target image.

[0088] Here, the rule-based algorithm calculates the average brightness value of the above electrode image according to predefined conditional instructions, selects at least one pixel with a difference greater than or equal to a certain range based on the calculated average brightness value, extracts a certain area so that the corresponding pixel is included to obtain an inspection target image, and can output this. For example, the inspection target image may be a binary image and can be provided in a size of 128 Pixel×128 Pixel.

[0089] FIG. 7 is a flowchart for explaining the step of determining the presence or absence of defects in an electrode among the battery electrode inspection methods according to an embodiment of the present invention.

[0090] Referring to FIG. 7, the battery electrode inspection apparatus 5000 according to an embodiment of the present invention can confirm the amount of learning data classified by a data set in the memory 100 (S5100). Here, at least one learning data can include at least one inspection result data pre-inspected by the battery electrode inspection apparatus 5000. In other words, the above learning data can include defective data or normal data. For example, the defective data can be classified and stored according to the defect type in the above data set.

[0091] Thereby, the battery electrode inspection apparatus 5000 can select the type of the already learned learning model.

[0092] According to an embodiment, when the amount of the learning data is less than a preset reference value, the battery electrode inspection apparatus 5000 can select the already learned first learning model (S5300).

[0093] Thereafter, the battery electrode inspection apparatus 5000 can extract an image feature value from the inspection target image (S5310).

[0094] Thereafter, the battery electrode inspection device 5000 can determine the presence or absence of defects in the battery electrode using the already learned first learning model (S5350).

[0095] On the other hand, according to another embodiment, when the amount of learning data is equal to or greater than a preset reference value, the battery electrode inspection device 5000 can select the already learned second learning model (S5500).

[0096] Thereafter, the battery electrode inspection device 5000 can determine the presence or absence of defects in the battery electrode using the selected second learning model (S5700).

[0097] FIG. 8 is a conceptual diagram for explaining a battery electrode inspection method using a first learning model among the battery electrode inspection methods according to an embodiment of the present invention.

[0098] Referring to FIG. 8, as described above, when the amount of learning data is less than a preset reference value, the battery electrode inspection device 5000 can select the already learned first learning model.

[0099] Thereafter, the battery electrode inspection device 5000 can extract an image feature value from the inspection target image.

[0100] Here, the image feature value may be at least one pixel information extracted from the inspection target image. For example, the features of the image feature value can include at least one of height, width, peak white, peak dark, aspect ratio, and circularity.

[0101] According to an embodiment, when the amount of learning data is less than a preset set value, the battery electrode inspection device 5000 can obtain an image feature value from at least one inspection target image using a rule-based algorithm.

[0102] After that, the battery electrode inspection device 5000 can determine whether there is a defect in the battery electrode by using the already learned first learning model.

[0103] More specifically, the battery electrode inspection device 5000 can input the image feature values extracted from the inspection target image into the already learned first learning model as input data. After that, the battery electrode inspection device 5000 can output result data regarding whether there is a defect in the battery electrode based on the inspection target image. Here, the battery electrode may be the electrode corresponding to the inspection target image.

[0104] According to an embodiment, the already learned first learning model may be a machine learning-based algorithm. For example, the first learning model may be a random forest of the bagging series that is pre-learned based on at least one learning data pre-classified in the data set in the memory 100.

[0105] FIG. 9 is a conceptual diagram for explaining a battery electrode inspection method using a second learning model among the battery electrode inspection methods according to an embodiment of the present invention.

[0106] Referring to FIG. 9, as described above, when the amount of learning data is equal to or greater than a preset reference value, the battery electrode inspection device 5000 can determine whether there is a defect in the battery electrode by using the already learned second learning model.

[0107] More specifically, the battery electrode inspection device 5000 can input the inspection target image into the already learned second learning model as input data. As a result, the battery electrode inspection device 5000 can output result data obtained by determining whether there is a defect in the battery electrode based on the inspection target image.

[0108] According to an embodiment, the already learned second learning model may be a pre-learned learning model based on at least one learning data pre-classified in the dataset in the memory 100.

[0109] More specifically described, the already learned second learning model may be an artificial intelligence (AI)-based classification model. In other words, the second learning model may be a deep-learning-based classification model pre-learned based on the already stored learning data based on the result data of the battery electrode inspection device 5000.

[0110] According to an embodiment, the already learned second learning model may be a learning model based on an artificial neural network of a convolutional neural network (CNN). For example, the battery electrode inspection device 5000 can extract image feature data of an inspection target image by a convolutional layer, reduce the feature data to a low dimension through a pooling layer, and classify it along the defect type in a fully connected layer, so that the second learning model can be pre-learned to determine the presence or absence of defects.

[0111] FIG. 10 is a conceptual diagram for explaining the step of re-learning among the battery electrode inspection methods according to another embodiment of the present invention.

[0112] Referring to FIG. 10, the battery electrode inspection device 5000 can store the result data obtained from one embodiment and another embodiment, in other words, the electrode defect presence or absence determination result data, in the memory 100 and classify it into a dataset.

[0113] Thereafter, the battery electrode inspection device 5000 can re-learn the learning model based on the updated at least one piece of learning data. In other words, the battery electrode inspection device 5000 can utilize the previously learned result data as learning data to re-learn the learning model. As a result, the amount of learning data increases, and the inspection reliability of the battery electrode inspection device 5000 can be improved.

[0114] As described above, the battery electrode inspection device and method according to the embodiments of the present invention have been explained.

[0115] The battery electrode inspection device and method according to the embodiments of the present invention extract an inspection target image including a region suspected of having a defect based on an electrode image acquired from a camera, and set and apply different types of learning models for determining the presence or absence of defects in the electrode corresponding to the inspection target image according to the amount of learning data, thereby shortening the application time due to the initial process application and process environment changes.

[0116] In addition, the battery electrode inspection device and method can obtain highly reliable result data with improved accuracy by re-learning the learning model based on the updated learning data.

[0117] The operations of the method according to the embodiments and experimental examples of the present invention can be embodied as a computer-readable program or code on a computer-readable recording medium. The computer-readable recording medium includes all types of recording devices in which data that can be read by a computer system is stored. In addition, the computer-readable recording medium can be distributed to a computer system connected via a network and store and execute a computer-readable program or code in a distributed manner.

[0118] In addition, a computer-readable recording medium can include a hardware device specially configured to store and execute program instructions, such as a read-only memory (ROM), a random access memory (RAM), a flash memory, etc. The program instructions can include not only machine language code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

[0119] Some aspects of the present invention have been described in the context of an apparatus, which can also represent corresponding descriptions by a corresponding method, where a block or apparatus corresponds to a method step or a feature of a method step. Similarly, aspects described in the context of a method can be represented by corresponding blocks or items or features of a corresponding apparatus. Some or all of the method steps can be performed (or used) by a hardware device such as a microprocessor, a programmable computer, or an electronic circuit. In some embodiments, one or more of the most important method steps can be performed by such a device.

[0120] As described above with reference to the preferred embodiments of the present invention, those skilled in the art will understand that the present invention can be variously modified and changed without departing from the spirit and scope of the present invention described in the following claims.

Description of Reference Numerals

[0121] 1000: Camera 5000: Battery Electrode Inspection Device 100: Memory 200: Processor 300: Transceiver 400: Input Interface Device 500: Output Interface Device 600: Storage Device 700: Bus

Claims

1. A battery electrode inspection device based on an artificial intelligence model, comprising a memory; and a processor that executes at least one instruction stored in the memory, wherein the at least one instruction includes an instruction to obtain an inspection target image by extracting a region suspected of being defective from an electrode image in which at least one electrode surface is photographed, and an instruction to input the inspection target image into a pre-trained learning model and output result data regarding the presence or absence of defects in the battery electrode. A battery electrode inspection device.

2. The instruction to output the result data includes an instruction to determine any one of a plurality of learning models according to at least one amount of learning data. The battery electrode inspection device according to claim 1.

3. The instruction to determine the learning model includes an instruction to output the result data using a pre-trained first learning model based on machine learning when the amount of learning data is less than a preset reference value. The battery electrode inspection device according to claim 2.

4. The instruction to determine the learning model includes an instruction to output the result data using a pre-trained second learning model based on deep learning when the amount of learning data is greater than or equal to a preset reference value. The battery electrode inspection device according to claim 3.

5. The instruction to output the result data using the first learning model includes an instruction to extract at least one image feature value from the inspection target image, and an instruction to input at least one of the image feature values into the first learning model and obtain the result data. The battery electrode inspection device according to claim 3.

6. The instruction to extract the image feature value includes an instruction to extract the image feature value from the inspection target image using a rule-based learning algorithm. The battery electrode inspection device according to claim 5.

7. The image feature value includes at least one data among the height of the pixel, the width of the pixel, the peak white value of the pixel, the peak dark value of the pixel, the aspect ratio, and the circularity ratio extracted from the inspection target image. The battery electrode inspection device according to claim 5.

8. The battery electrode inspection device according to claim 3, wherein the first learning model is a learning model based on Random Forest.

9. The instruction to output the result data using the second learning model includes an instruction to input at least one inspection target image into the second learning model to obtain the result data. The battery electrode inspection device according to claim 4.

10. The battery electrode inspection device according to claim 4, wherein the second learning model is a learning model based on a Convolutional Neural Network (CNN).

11. The at least one instruction further includes an instruction to relearn the already learned learning model by utilizing the result data as learning data. The battery electrode inspection device according to claim 1.

12. A battery electrode inspection method using an artificial intelligence model-based battery electrode inspection device, comprising: obtaining an inspection target image by extracting a region suspected of having a defect from an electrode image in which at least one electrode surface is photographed; and inputting the inspection target image into an already learned learning model and outputting result data regarding the presence or absence of a defect in the battery electrode. The battery electrode inspection method.

13. The step of outputting the result data includes a step of determining any one of a plurality of learning models according to at least one amount of learning data. The battery electrode inspection method according to claim 12.

14. The step of determining the learning model includes an instruction to output the result data using a first learning model based on already learned machine learning when the amount of learning data is less than a preset reference value. The battery electrode inspection method according to claim 13.

15. The step of determining the learning model includes a step of outputting the result data using a second learning model based on already learned deep learning when the amount of learning data is greater than or equal to a preset reference value. The battery electrode inspection method according to claim 14.

16. The step of outputting the result data using the first learning model includes an instruction to extract at least one image feature value from the inspection target image, and The battery electrode inspection method according to claim 14, comprising the step of inputting at least one of the image feature values into the first learning model to obtain the result data.

17. The step of extracting the image feature values The battery electrode inspection method according to claim 16, comprising the step of extracting the image feature values from the inspection target image using a rule-based learning algorithm.

18. The image feature values The battery electrode inspection method according to claim 16, including at least one data among the height of the pixel, the width of the pixel, the peak white value of the pixel, the peak dark value of the pixel, the aspect ratio, and the circularity ratio extracted from the inspection target image.

19. The first learning model is a learning model based on random forest, and the battery electrode inspection method according to claim 14.

20. The step of outputting the result data using the second learning model The battery electrode inspection method according to claim 15, including the step of inputting at least one inspection target image into the second learning model to obtain the result data.

21. The second learning model is a learning model based on a convolutional neural network (CNN), and the battery electrode inspection method according to claim 15.

22. The battery electrode inspection method The battery electrode inspection method according to claim 12, further comprising the step of using the result data as learning data to relearn the already learned learning model.

23. An artificial intelligence model-based battery electrode inspection system, comprising A camera that captures at least one electrode surface to generate an electrode image; and A battery electrode inspection system including a battery electrode inspection device that obtains an inspection target image by extracting a region suspected of being defective from the electrode image, inputs the inspection target image into a already learned learning model, and outputs result data regarding the presence or absence of defects in the battery electrode.

Images