The automatic detection of fabric defects based on computer vision by using adaptive wavelet transform was studied in this paper. A new method of constructing the adaptive orthonormal wavelet basis by considering the orthonormal condition and constrain condition respectively is proposed. It includes to estabalish an orthonormal wavelet database for a certain length of wavelet bases according to orthonormal condition and to optimize the wavelet bases by taking the energy of high-pass decomposed subimage to be minimum as the constraint condition, i. e. to search out the optimum bases from the database by genetic algorithm to adapt to the fabric texture. An adaptive wavelet decomposition of fabric images with one resolution level is also proposed. The feature parameters extraction and window section in the decomposed images are described in detail. Five different feature parameters such as energy, standard deviation, entropy, extreme deviation and contrast are considered to be the indexes which can multiply manifest the abnormal change of the fabric texture. At last the automatic inspection procedures to detect fabric defects are introduced and the test results of the fabric defects such as slub yarn, double pick, miss pick, miss warp yarn, double warp, tight thread, knees, fabric hole, gout, weft crackiness, oiled defect and etc are discussed.
展开▼