Tyre Defect Detection Based on GLCM and Gabor Filter

Abstract

Defect detection is a crucial issue at industrial level, and generally, the designers have to find out a way to select suitable features to detect the defects in textured material images. This work reports this issue using the gray-level co-occurrence matrix (GLCM) and Gabor filter texture segmentation. We first analyze the defects in the corresponding tyre texture images using GLCM and then we apply Gabor filter texture segmentation to illustrate further the impurities found in the material. Our proposed method matches the abnormal textured images with standard texture images, and we use the 5 GLCM features to calculate the score for an abnormal texture image. Further, our proposed method decides the defected tyres and presents the texture segmentation of abnormal image. We found that our proposed method performs well when compares with other methods of tyre images data.