Published Paper Details:

There are different applications of computer vision and digital image processing in various applied domains and automated production processes. Automatic defect detection is crucial in the textile industry as the quality and price of textile products rely on its efficiency and effectiveness. Fabric defect detection has been a challenging task, and manual human efforts were previously used to detect defects in the fabric production process. The major drawbacks of the manual fabric defect detection method are lack of concentration, human fatigue, and time consumption. Applications based on computer vision and digital image processing can handle the aforementioned limitations and drawbacks. Numerous computer vision-based applications have been proposed in research articles over the past two decades to overcome these limitations. The objective of this review article is to provide a comprehensive analysis of different computer vision-based techniques that are applied in the textile industry for fabric defect detection. The suggested study presents an extensive analysis of various techniques such as histogram-based methods, color-based methods, image segmentation-based methods, frequency domain operations, texture-based defect detection, sparse feature-based operations, image morphology operations, and recent advancements in deep learning. The performance evaluation criteria for automatic fabric defect detection are also presented and discussed. The disadvantages and limitations of the existing published study are thoroughly discussed, as are potential future research directions. This research study offers in-depth information on computer vision and digital image processing applications for detecting various types of fabric defects.