An Automated Surface Defect Inspection System Using Local Binary Patterns and Co-Occurrence Matrix Textures based on SVM Classifier

Computer vision techniques have been widely used in automating the surface defect inspection process where the goal is to detect and identify defects. surface defect is defined as global color deviation or local textural irregularity which has the main concern in the inspection process. in this paper, the proposed automated system identifies different surface defects using support vector machine (svm) classifier according to surface textural features. the proposed system introduces a novel feature description technique that extracts local and global features of surfaces. this technique combines local binary pattern (lbp) features with the global textural features of gray-level co-occurrence matrix (glcm) to address different surface defects. the proposed system has been tested on wood and ceramic tiles images. experimental results successfully demonstrated the efficiency of the feature description technique and the overall surface defect inspection system.