Image Texture Defect Detection Method Using Fuzzy C-Means Clustering for Visual Inspection Systems.

A new texture defect detection method for automatic visual inspection systems is presented in this paper. It divides an analysed texture image into non-overlapping samples, and then calculates features of each sample using the Principle Component Analysis technique. Finally, the fuzzy c-means clustering of these features is applied to classify the sample as defective or non-defective. Unlike many existing methods, the proposed scheme does not require a training step to collect defective and non-defective texture samples. The experimental results show that the method is at least as effective and accurate as many existing methods for image texture defect detection. [ABSTRACT FROM AUTHOR]