A Coupling Enhancement Algorithm for ZrO2 Ceramic Bearing Ball Surface Defect Detection Based on Cartoon-texture Decomposition Model and Multi-Scale Filtering Method

To improve the detection accuracy of surface defects of ZrO2 ceramic bearing balls, a detection method for surface defects of ZrO2 ceramic bearing balls based on cartoon texture decomposition model is proposed. By building a surface defect detection system for ZrO2 ceramic bearing balls, the Gaussian curvature model is used to decompose the surface defect images of ZrO2 ceramic bearing balls, and the decomposed image layers are filtered by winner filtering and wavelet value domain filtering, and fused into clear and damage-free images The surface defect image of ZrO2 ceramic bearing ball and its detection. The experimental results shown the PSNR of image is 34.1 dB, the SSIM is 0.9476, the detection accuracy is 95.8 %, and the detection speed of a single defect image is 191 ms/img. The method can remove images noise, retain its details. It improves the efficiency and accuracy of ZrO2 ceramic bearing ball surface defect detection.