Handloom Silk Fabric Defect Detection Using First Order Statistical Features on a NIOS II Processor

This paper focuses on identifying defects in a handloom silk fabric using image analysis techniques such as first order statistical features. Any disparity in the knitting process that leads to an unpleasant appearance or dissatisfaction of the customer is termed as a defect in the fabric. Even today, the defect detection in a silk fabric is done using skilled manual labour. An automated defect detection and identification system would naturally enhance the quality and result in improved productivity to meet both customer demands and also reduce the costs associated with off-quality. This paper also classifies about the various defects that can occur in a silk fabric. As a supplementary need for the proposed machine vision based defect detection in textile fabric images, we would require a hardware implementation of the proposed method. This has been done using a soft core processor such as a NIOS processor of Altera Semiconductors.