Lithium Battery Pole Piece Defect Detection Method Based on Mean Shift and Gray-level Co-occurrence Matrix

In response to the problems of low efficiency and low accuracy of the traditional manual method of detecting defects in lithium battery poles. In this paper, we propose a way to detect the defects of lithium battery poles based on the combination of mean shift and gray-level co-occurrence matrix (GLCM). Firstly, ROI extraction of the coated area of the lithium battery pole piece, noise reduction using the mean drift algorithm, and image contrast enhancement using the Otsu method. Then, the defect segmentation is performed using the gray-level co-occurrence matrix. Finally, traverse the edge pixels and use the minimum outer rectangle method for defect contouring. The results show that the method in this paper can obtain good detection results for different types of defects, such as decarburization, metal leakage, material dropout, and white spots, which are of practical value to the lithium battery pole pieces manufacturing industry.