Fabric quality testing using image processing

Abstract

The main purpose of this paper is to identify the damaged cloth which may contain defective yarns, colour bleeding, and pores that may be mingled with the good cloth materials at areas such as textile fabric industries, garments, and weaving factories. Automated defect detection for fabrics based on filters is proposed to manage the problem of human visual inspection. The monitoring process is done by a web camera, to capture the details of the cloth. A pre-trained Gabor wavelet network is used to extract the important texture features in the textile fabric. In this defect detection, few specific filters are used to process each frames of the cloth which is being captured by the web camera.