Automatic Detection of Cloth Defects based on Laws Texture Filtering
Download as PDF
Guanxiong Ding, Hui Huang, and Yan Ma
In this paper, the difference between the fabric features of the cloth and the normal part is used to realize the automatic detection of the flaws in the cloth. And the classic Laws texture filtering method and support vector machine (SVM) classifier are used to detect the flaws. The detection mainly includes two steps. The first step is to divide the cloth image into multiple sub-blocks of the same size, each sub-block as a sample, and then calculate the texture features of each sub-block separately, and the second step uses SVM classifier to different the samples of cloths. In the second step, the SVM classifier is used to classify different cloth samples to detect the fabric defects. Experiments show that the proposed algorithm can basically realize the automatic detection function of flaws in the cloth and can effectively identify whether the cloth has flaws.
Flaw detection, Laws filter, texture energy, Support Vector Machine