A novel surface defect detection method of cold rolled strips based on Artificial Immune System

Because surface defect image of cold rolled strips is disturbed by mass of noise information, as well as the image quality problem of inadequate illumination or uneven illumination, it will raise great difficulty when detecting the defect with traditional image process methods such as mathematic morphology, and it cannot get an ideal treatment effect. According to the problem, and combining the self-organizing and self-recognition features of Artificial Immune System technology, surface defect detection method of cold rolled strips based on AIS is studied. By assuring the including relationships among detectors and antigens, and the position information of self-body in domain space, the block space mode is introduced to present a block generating algorithm based on detector, and applied to detect surface defect image of cold rolled strips. Experiments show that this method is significantly superior in defect extraction to traditional defect image detection algorithms not only for images with low-contrast but also for images with inadequate illumination or uneven illumination.