Defect Detection in Textiles Using Morphological Analysis of Optimal Gabor Wavelet Filter Response

This paper addresses new defect detection method in textile based on Morphological Analysis and Gabor wavelet filters responses. Our method consist of there part. First a bank of Gabor wavelet filters is applied on the textile image for extracting feature matrix. It is based on the energy response from the convolution of Gabor wavelet filters in different frequency and orientation domains. Then based on the response of filters, the optimal filter is selected among the filter bank. Considering the industrial requirement for finding adaptive solutions that can be executed in real time and reduce false rate, we use morphological analysis as an adaptive threshold technique for detection. By applying morphological analysis on response of the optimal filter, the defects are detected. The experimental results on different type of textiles show that the developed algorithm is robust, scalable and effective for detection various kind of textile defects.