Fabric Defect Detection using Image Processing

Fabric defect is one of the most important and serious matters of quality control in textile industry in Bangladesh. This task takes a lot of time and money. For this reason we have introduced a simple process to find defects on fabric based on edge detection. This process is mainly focused on image processing which can be integrated with fabric defect detection automation system. In this paper we have tried a new approach using the filter method with edge detection and found good results. Our algorithm can detect defected fabric area successfully. It can be also used in real-time defect detection considering light intensity, zoom, fabric width, camera resolution etc. As our algorithm mainly works on the principle of edge detection, it cannot detect defect on multicoloured or patterned fabric. It works well on single coloured fabric without any fold or edge.