Lithium battery production quality inspection system based on visual inspection

By combining X-ray and camera imaging technologies, the lithium battery structure image is segmented, corrected, and compared, thus solving the problem of the impact of changes in ambient light on detection accuracy and achieving efficient quality control in lithium battery production.

CN119000710AInactive Publication Date: 2024-11-22山东齐创石化工程有限公司
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411106331.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing visual inspection systems are easily affected by changes in ambient light during lithium battery production, leading to decreased inspection accuracy, misjudgments, and the entry of some defective products into the market.

Method used

The method combines X-ray imaging and camera imaging to acquire images of the lithium battery structure. The image information is then segmented, numbered, corrected, and compared using an image preprocessing and analysis module. An algorithm is then used to determine whether the lithium battery has defects.

Benefits of technology

This improved the accuracy of lithium battery testing, avoided misjudgments, ensured that defective products were detected and handled in a timely manner, and improved production quality control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119000710A_ABST
    Figure CN119000710A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of lithium battery production quality inspection, and discloses a lithium battery production quality inspection system based on visual inspection, which comprises an image recognition module, an image data preprocessing module, an image data analysis module and a judgment module. According to the lithium battery production quality inspection system based on visual inspection, the data are compared with reference data, whether errors exist or not is judged, when any one of the two kinds of data is abnormal, it is judged that the lithium battery structure has small abnormity, namely small defects exist, a corresponding signal is sent out, and the lithium battery production quality inspection system is operated. When abnormal data of the two pictures simultaneously have large abnormity, namely large defects exist, a corresponding signal is sent out, and the judgment module displays the number of a defect area to prompt a worker to replace a structure corresponding to the lithium battery or repair the defects on the structure, so that the accuracy of a detection result is improved, and the detection efficiency is improved. The phenomenon that some defects cannot be found, misjudgment is caused, and products with the defects appear to flow into the market is avoided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of lithium battery production quality inspection, and in particular to a lithium battery production quality inspection system based on visual inspection. Background Art

[0002] The visual inspection system for lithium battery production quality inspection uses computer vision technology to automatically detect defects and abnormalities in the production process of lithium batteries. Its core is to monitor and analyze the battery in real time through cameras and image processing algorithms to ensure that product quality and performance meet standards. At present, it is usually based on cameras to take pictures of the lithium battery structure and compare it with samples to determine whether there are defects in the structure. However, relying solely on cameras to identify defects is easily affected by changes in ambient lighting conditions, which affects image quality and thus affects detection accuracy, resulting in some defects not being discovered, leading to misjudgment, and causing defective products to enter the market. Summary of the invention

[0003] 1. Technical issues to be solved

[0004] In view of the deficiencies in the prior art, the present invention provides a lithium battery production quality inspection system based on visual inspection, which has advantages such as *** and solves the above problems.

[0005] (II) Technical solution

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a lithium battery production quality inspection system based on visual inspection, comprising an image recognition module, an image data preprocessing module, an image data analysis module and a determination module;

[0007] The image recognition module is composed of an X-ray imaging acquisition unit and a camera imaging acquisition unit. The X-ray imaging acquisition unit is used to acquire battery structure image information A, and the camera imaging acquisition unit is used to acquire battery structure image information A′. The image recognition module sends the acquired cell structure image information A and battery structure image information A′ to the image preprocessing module;

[0008] The image preprocessing module divides the received pool structure image information A and battery structure image information A' into several regions and uniquely numbers each region. The several regions of the battery structure image information A are numbered as: a, b, c, ..., n, and the several regions of the battery structure image information A' are numbered as: a', b', c', ..., n'. The image preprocessing module corrects the image information of each region to obtain the regional corrected image information X and the regional corrected image information X', respectively. The image data preprocessing module sends the regional corrected image information X and the regional corrected image information X' to the image data analysis module;

[0009] The image data analysis module is respectively preset with X image samples and camera image samples, the X image samples are divided into several reference areas and correspondingly contain several reference data CX, the camera image samples are divided into several reference areas and correspondingly contain several reference data CX′, the image analysis module respectively brings the regional correction image information X, the regional correction image information X′, the reference data CX and the reference data CX′ into the algorithm formula to calculate and judge whether the lithium battery has defects, and sends a corresponding signal to the judgment module;

[0010] After receiving the signal, the determination module determines and displays whether the lithium battery is qualified.

[0011] Preferably, the area corrected image information X and the area corrected image information X′ correspond to the reference data CX and the reference data CX′ respectively, and the several areas of the battery structure image information A, the several areas of the battery structure image information A′, the several reference areas of the X image sample and the several reference areas of the camera image sample are divided one by one according to the image coordinates.

[0012] Preferably, the region corrected image information X includes coordinate data (X n ,Y n ) and the pixel value within the coordinates, the regional correction image information X′ includes the coordinate data (X′) of the region n , Y′ n ) and the pixel value within the coordinates, the reference data CX includes the coordinate data of the area (KX n ,KY n ) and the pixel value within the coordinates, the reference data CX′ includes the coordinate data of the area (KX′ n , KY′ n ) and the pixel values ​​within the changed coordinates.

[0013] Preferably, the algorithm expression of the region corrected image information X is as follows:

[0014] X=CDF(i(X n ,,Y n ))*(L-1)-1

[0015] In the formula, CDF(i(X n ,,Y n )) represents the pixel value i(X n ,,Y n ), i corresponds to the pixel data of the corresponding coordinates of several areas a, b, c, ..., n of the battery structure image information A, L-1 represents the maximum grayscale value, and -1 represents the adjustment of the image format.

[0016] Preferably, the algorithm expression of the region corrected image information X′ is as follows:

[0017] X′=CDF(i′(X′ n , Y′ n ))*(L-1)-1

[0018] In the formula, CDF(i′(X′ n , Y′ n )) represents the pixel value i′(X′) in the image n , Y′ n ) Cumulative distribution function, i′ corresponds to the pixel data of the corresponding coordinates of several areas a′, b′, c′, ..., n′ of the battery structure image information A′, L-1 represents the maximum grayscale value, and -1 represents the adjustment of the image format.

[0019] Preferably, the algorithm formula for comparing the regional corrected image information X with the reference data CX is as follows:

[0020] Cz(X n ,Y n )=|X(X n ,Y n )-CX(KX n ,KY n )|

[0021] In the formula, Cz(X n ,Y n ) represents the difference between the pixel value in the area correction image information X and the pixel value in the reference data CX in the same coordinate area, |X(X n ,Y n )-CX(KX n ,KY n )| means the difference result is taken as the absolute value.

[0022] Preferably, the algorithm formula for comparing the regional corrected image information X′ with the reference data CX′ is as follows:

[0023] Cz′(X n ,Y n )=|X′(X′ n , Y′ n )-CX′X′ n , Y′ n )|

[0024] In the formula, Cz′(X n ,Y n ) represents the difference between the pixel value in the area correction image information X′ and the pixel value in the reference data CX′ in the same coordinate area, |X′(X′ n , Y′ n)-CX′(X′ n , Y′ n )| means the difference result is taken as the absolute value.

[0025] Preferably, the image data analysis module calculates Cz(X n ,Y n ), when Cz(X n ,Y n ) is greater than 0, it is determined that the battery structure image information A is abnormal, and the image data analysis module calculates Cz′(X n ,Y n ),when Cz′(X n ,Y n ) is greater than 0, it is determined that the battery structure image information A′ is abnormal, an abnormal signal 1 or an abnormal signal 2 is generated, and the two signals are sent to the determination module.

[0026] Preferably, when the determination module receives only abnormal signal 1 or abnormal signal 2, the determination module determines that the defect in the lithium battery structure is small and displays the number of the defective area.

[0027] Preferably, when the determination module receives abnormal signal 1 and abnormal signal 2 simultaneously, the determination module determines that there is a large defect in the lithium battery structure and displays the number of the defective area.

[0028] Compared with the prior art, the present invention provides a lithium battery production quality inspection system based on visual inspection, which has the following beneficial effects:

[0029] The present invention collects images generated by X-rays and images collected by cameras respectively, divides the collected images into areas according to coordinates, and corrects the image information in each area. After the accuracy of the collected image information is increased, it is compared with reference data to determine whether there is an error. When any one of the two images has an abnormality, it is determined that there is a small abnormality in the lithium battery structure, that is, there is a small defect, and a corresponding signal is sent. When abnormal data appears in both images at the same time, there is a large abnormality, that is, there is a large defect, and a corresponding signal is sent. The number of the defective area is displayed through the judgment module to prompt the worker to replace the corresponding structure of the lithium battery or repair the defects on the structure, thereby improving the accuracy of the detection result, avoiding the phenomenon of misjudgment caused by the inability to discover some defects, and causing the product with abnormality to enter the market. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the system flow of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] See also Figure 1 , a lithium battery production quality inspection system based on visual inspection, including an image recognition module, an image data preprocessing module, an image data analysis module and a determination module;

[0033] The image recognition module is composed of an X-ray imaging acquisition unit and a camera imaging acquisition unit. The X-ray imaging acquisition unit is used to acquire battery structure image information A, and the camera imaging acquisition unit is used to acquire battery structure image information A′. The image recognition module sends the acquired cell structure image information A and battery structure image information A′ to the image preprocessing module.

[0034] The image preprocessing module divides the received pool structure image information A and battery structure image information A′ into a plurality of regions and uniquely numbers each region. The plurality of regions of the battery structure image information A are numbered as: a, b, c, ..., n, and the plurality of regions of the battery structure image information A′ are numbered as: a′, b′, c′, ..., n′;

[0035] The image preprocessing module corrects the image information of each region to obtain the region corrected image information X and the region corrected image information X′, respectively. The image data preprocessing module sends the region corrected image information X and the region corrected image information X′ to the image data analysis module, wherein:

[0036] The area correction image information X includes the coordinate data of the area (X n ,Y n ) and the pixel value within the coordinates, the regional correction image information Y′ includes the coordinate data (X′) of the region n , Y′ n ) and the pixel value within that coordinate;

[0037] The algorithm expression for calculating the region correction image information X and the region correction image information X′ is as follows:

[0038] X=CDF(i(Y n ,,Y n ))*(L-1)-1

[0039] In the formula, CDF(i(X n ,,Y n )) represents the pixel value i(Xn ,,Y n ), i corresponds to the pixel data of the corresponding coordinates of several areas a, b, c, ..., n of the battery structure image information A, L-1 represents the maximum grayscale value, and -1 represents the adjustment of the image format;

[0040] The algorithm expression of regional correction image information X′ is as follows:

[0041] X′=CDF(i′(X′ n , Y′ n ))*(L-1)-1

[0042] In the formula, CDF(i′(X′ n , Y′ n )) represents the pixel value i′(X′) in the image n , Y′ n ) Cumulative distribution function, i′ corresponds to the pixel data of the corresponding coordinates of several areas a′, b′, c′, ..., n′ of the battery structure image information A′, L-1 represents the maximum grayscale value, and -1 represents the adjustment of the image format.

[0043] After the regional correction image information X and the regional correction image information X′ are calculated, the image data preprocessing module sends the regional correction image information X and the regional correction image information X′ to the image data analysis module, wherein the image data analysis module is preset with X image samples and camera image samples, the X image samples are divided into several reference areas, and correspondingly contain several reference data CX, the camera image samples are divided into several reference areas, and correspondingly contain several reference data CX′, it should be noted that the regional correction image information X and the regional correction image information X′ correspond to the reference data CX and the reference data CX′ respectively, and the several areas of the battery structure image information A, the several areas of the battery structure image information A′, the several reference areas of the X image samples and the several reference areas of the camera image samples are divided one by one according to the image coordinates;

[0044] The reference data CX includes the coordinate data of the area (KX n ,KY n ) and the pixel value within the coordinates, the reference data CX′ includes the coordinate data of the area (KX′ n , KY′ n ) and the pixel value within the changed coordinates;

[0045] The image analysis module respectively brings the regional correction image information X, the regional correction image information X′, the reference data CX and the reference data CX′ into the algorithm formula to calculate and judge whether the lithium battery has defects, and sends a corresponding signal to the judgment module. The specific comparison algorithm is as follows:

[0046] The algorithm formula for comparing the regional corrected image information X with the reference data CX is as follows:

[0047] Cz(X n ,Y n )=|X(X n ,Y n )-CX(KX n ,KY n )|

[0048] In the formula, Cz(X n ,Y n ) represents the difference between the pixel value in the area correction image information X and the pixel value in the reference data CX in the same coordinate area, |X(X n ,Y n )-CX(KX n ,KY n )| means the difference result takes the absolute value;

[0049] The algorithm formula for comparing the regional corrected image information X′ with the reference data CX′ is as follows:

[0050] Cz′(X n ,Y n )=|X′(X′ n , Y′ n )-CX′(X′ n , Y′ n}|

[0051] In the formula, Cz′(X n ,Y n ) represents the difference between the pixel value in the area correction image information X′ and the pixel value in the reference data CX′ in the same coordinate area, |X′(X′ n , Y′ n )-CX′(X′ n , Y′ n )| means the difference result takes the absolute value;

[0052] The image data analysis module calculates Cz(X n ,Y n ), when Cz(X n ,Y n ) is greater than 0, it is determined that the battery structure image information A is abnormal, and the image data analysis module calculates Cz′(X n ,Y n ),when Cz′(X n ,X n ) is greater than 0, it is determined that the battery structure image information A′ is abnormal, an abnormal signal 1 or an abnormal signal 2 is generated, and the two signals are sent to the determination module;

[0053] When the module receives the signal, it determines and displays whether the lithium battery is qualified. The specific determination steps are as follows:

[0054] 1. When the determination module receives only abnormal signal 1 or abnormal signal 2, the determination module determines that the defect in the lithium battery structure is small and displays the number of the defective area;

[0055] 2. When the determination module receives abnormal signal 1 and abnormal signal 2 at the same time, the determination module determines that there is a major defect in the lithium battery structure and displays the number of the defective area;

[0056] When the defective area number is displayed, the worker repairs or replaces the defective part of the lithium battery according to the displayed area number;

[0057] By collecting images generated by X-rays and images collected by cameras respectively, dividing the collected images into areas according to coordinates, and correcting the image information in each area, the accuracy of the collected image information is increased, and then it is compared with the reference data to determine whether there is an error. When any one of the two images has an abnormality, it is determined that there is a small abnormality in the lithium battery structure, that is, there is a small defect, and a corresponding signal is sent. When abnormal data appears in both images at the same time, there is a large abnormality, that is, there is a large defect, and a corresponding signal is sent. The number of the defective area is displayed through the judgment module to prompt the worker to replace the corresponding structure of the lithium battery or repair the structural defects, thereby improving the accuracy of the detection results, avoiding the failure to detect some defects, resulting in misjudgment, and causing the product with defects to enter the market.

[0058] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A lithium battery production quality inspection system based on visual inspection, characterized by: It includes an image recognition module, an image data preprocessing module, an image data analysis module and a determination module; The image recognition module is composed of an X-ray imaging acquisition unit and a camera imaging acquisition unit. The X-ray imaging acquisition unit is used to acquire battery structure image information A, and the camera imaging acquisition unit is used to acquire battery structure image information A′. The image recognition module sends the acquired cell structure image information A and battery structure image information A′ to the image preprocessing module; The image preprocessing module divides the received pool structure image information A and battery structure image information A' into several regions and uniquely numbers each region. The several regions of the battery structure image information A are numbered as: a, b, c, ..., n, and the several regions of the battery structure image information A' are numbered as: a', b', c', ..., n'. The image preprocessing module corrects the image information of each region to obtain the regional corrected image information X and the regional corrected image information X', respectively. The image data preprocessing module sends the regional corrected image information X and the regional corrected image information X' to the image data analysis module; The image data analysis module is respectively preset with X image samples and camera image samples, the X image samples are divided into several reference areas and correspondingly contain several reference data CX, the camera image samples are divided into several reference areas and correspondingly contain several reference data CX′, the image analysis module respectively brings the regional correction image information X, the regional correction image information X′, the reference data CX and the reference data CX′ into the algorithm formula to calculate and judge whether the lithium battery has defects, and sends a corresponding signal to the judgment module; After receiving the signal, the determination module determines and displays whether the lithium battery is qualified.

2. A lithium battery production quality inspection system based on visual inspection according to claim 1, characterized in that: The area corrected image information X and the area corrected image information X′ correspond to the reference data CX and the reference data CX′ respectively, and the several areas of the battery structure image information A, the several areas of the battery structure image information A′, the several reference areas of the X image sample and the several reference areas of the camera image sample are divided one by one according to the image coordinates.

3. A lithium battery production quality inspection system based on visual inspection according to claim 2, characterized in that: The area correction image information X includes the coordinate data (X n , Y n ) and the pixel value within the coordinates, the regional correction image information X′ includes the coordinate data (X′) of the region n , Y′ n ) and the pixel value within the coordinates, the reference data CX includes the coordinate data of the area (KX n ,KY n ) and the pixel value within the coordinates, the reference data CX′ includes the coordinate data of the area (KX′ n , KY′ n ) and the pixel values ​​within the changed coordinates.

4. The lithium battery production quality inspection system based on visual inspection according to claim 3 is characterized by: The algorithm expression of the region correction image information X is as follows: X=CDF(i(X n ,,Y n ))*(L-1)-1 In the formula, CDF(i(X n ,,Y n )) represents the pixel value i(X n ,,Y n ), i corresponds to the pixel data of the corresponding coordinates of several areas a, b, c, ..., n of the battery structure image information A, L-1 represents the maximum grayscale value, and -1 represents the adjustment of the image format.

5. The lithium battery production quality inspection system based on visual inspection according to claim 3, characterized in that: The algorithm expression of the regional correction image information X′ is as follows: X′=CDF(i′(X′ n ,Y′ n ))*(L-1)-1 In the formula, CDF(i′(X′ n , Y′ n )) represents the pixel value i′(X′) in the image n , Y′ n ) Cumulative distribution function, i′ corresponds to the pixel data of the corresponding coordinates of several areas a′, b′, c′, ..., n′ of the battery structure image information A′, L-1 represents the maximum grayscale value, and -1 represents the adjustment of the image format.

6. A lithium battery production quality inspection system based on visual inspection according to claim 5, characterized in that: The algorithm formula for comparing the regional corrected image information X with the reference data CX is as follows: Cz(X n ,Y n )=|X(X n ,Y n )-CX(KX n ,KY n )| In the formula, Cz(X n , Y n ) represents the difference between the pixel value in the area correction image information X and the pixel value in the reference data CX in the same coordinate area, |X(X n , Y n )-CX(KX n ,KY n )| means the difference result is taken as the absolute value.

7. A lithium battery production quality inspection system based on visual inspection according to claim 6, characterized in that: The algorithm formula for comparing the regional corrected image information X′ with the reference data CX′ is as follows: Cz′(X n ,Y n )=|X′(X′ n ,Y′ n )-CX′(X′ n ,Y′ n )| In the formula, Cz′(X n , Y n ) represents the difference between the pixel value in the area correction image information X′ and the pixel value in the reference data CX′ in the same coordinate area, |X′(X′ n , Y′ n )-CX′(X′ n , Y′ n )| means the difference result is taken as the absolute value.

8. The lithium battery production quality inspection system based on visual inspection according to claim 7, characterized in that: The image data analysis module calculates Cz(X n , Y n ), when Cz(X n , Y n ) is greater than 0, it is determined that the battery structure image information A is abnormal, and the image data analysis module calculates Cz′(X n , Y n ),when Cz′(X n , Y n ) is greater than 0, it is determined that the battery structure image information A′ is abnormal, an abnormal signal 1 or an abnormal signal 2 is generated, and the two signals are sent to the determination module.

9. A lithium battery production quality inspection system based on visual inspection according to claim 8, characterized in that: When the determination module receives only abnormal signal 1 or abnormal signal 2, the determination module determines that the defect in the lithium battery structure is small and displays the number of the defective area.

10. A lithium battery production quality inspection system based on visual inspection according to claim 9, characterized in that: When the determination module receives abnormal signal 1 and abnormal signal 2 simultaneously, the determination module determines that there are large defects in the lithium battery structure and displays the number of the defective area.

Citation Information

Patent Citations

  • Method and device for automatically detecting internal defects of lithium battery cell

    CN113129260A

  • Hub internal defect detection method, device, equipment, medium and program product

    CN113470024A

  • Lithium battery defect detection method and device based on machine vision and electronic equipment

    CN116519707A