Fabric inspection based on best wavelet packet bases

In this study, we use best wavelet packet bases and an artificial neural ne twork (ANN) to inspect four kinds of fabric defects. Multiresolution repres entation of an image using wavelet transform is a new and effective approac h for analyzing image information content. In this study, we find the value s and positions for the smallest-six entropy in a wavelet packet best tree that acts as the feature parameters of the ANN for identifying fabric defec ts. We explore three basic considerations of the classification rate of fab ric defect inspection comprising wavelets with various maximum vanishing mo ments, different numbers of resolution levels, and differently scaled fabri c images. The results show that the total classification rate for a wavelet function with a maximum vanishing moment of four and three resolution leve ls can reach 100%, and differently scaled fabric images have no obvious eff ect on the classification rate.