Fabric Defect Detection Image Processing Technology

The detection of fabric defects by digital image processing and analyzing technique was studied in this thesis. The main contents were as follows: the status quo in the research of detection of fabric defects, the hardware and software environment of the detection of fabric defects, the classification of the common fabric defects, and image segmentation and feature extraction of various defects, etc. The followings are the brief introductions of the contents for each chapter:In Chapter 1, the background of the topic of this thesis and the main contents to study in this paper were introduced.In Chapter 2, a new classification method of fabric defects was developed. Hundreds of fabric defects were briefly divided into four types: warp defects, weft defects, local defects and discrete defects. Besides the hardware and software environment constituting fabric defects detection was introduced.In Chapter 3, image preprocessing and segmentation of the four types of defects were introduced. And these image preprocessing and segmentation were applied in image data of various fabric defects, which were captured under the same conditions. After separating defects from the texture background, we got the desired images of fabric defects.In Chapter 4, the geometric meaning of the chosen characteristic was presented firstly. Then effective characteristic parameters were extracted from different defects. Finally, we carried out a comprehensive analysis of the features of all kinds of defects, and succeeded to represent geometric features of four types of defects in a graph, which helps to further pattern recognition.In Chapter 5, efforts of the whole thesis were summarized, then disadvantages were described and suggestions for further study were put forward.