Localization of Defects in Fabric Imagery Using Contrast Limited Adaptive Histogram Equalization and Mathematical Morphology

Automated visual inspection systems are a long felt alternative to human visual inspection systems in the textile industry, especially when the quality control of products in the industry is a significant problem. It is observed that the price of textile fabric is reduced by 45% to 65% due to defects. In the manual fault detection systems with trained inspectors, only very few defects are being detected while an automatic system can increase this to a maximum number thus, automated visual inspection systems play a great role in assessing the quality of textile fabrics. For the detection of defects in a homogeneous fabric, we first perform bit plane decomposition of the image. The lower order bit planes are found to be useful for the localization of defects while eliminating the prints and other parts. Then we extract the exact boundary by means of mathematical morphology. The algorithm has been evaluated on a subset of TILDA1 image database with various visual qualities. Robustness with respect to the changes of the parameters of the algorithm has been examined.