Surface Defects Detection Techniques For Steel Strips Based On Artificial Immune System

Though online surface inspection systems for cold rolled strips have been widely used, some essential problems were not solved effectively, such as low detection rate for strip surface low contrast defects (Bond defects in low-contrast, Defects in low-contrast printing roll, etc), a high degree of information on strip surface defects (Hole defects, Scratch defects, etc). In addition, with domestic production of cold rolled strip steel sector to gradually increase the rolling speed, real-time processing performance of online surface inspection systems for cold rolled strips will be adapted to these changes.Strip surface defect detection technology has the striking similarities as the working principle of the immune system. We use it to research and design the strip surface defects detection technology based on artificial immune system, and realize the defects detection technology with high efficiency, thus can achieve quality of steel strip surface defect detection.Main works and achievements are as follows:1. In order to realize detection techniques of the strip surface,according to this problem, a new strip surface defects detection technology is proposed negative selection algorithm based on B tree. According to factors of generation of mature detectors, the candidate detector are produced variation through low frequency and high frequency, representation of detector is changed to value-based detector instead of binary-based detector, and 3 properties such as self-number, detector-number and some antigen-number are extracted to construct indexes using B tree based negative selection algorithm to achieve quality of steel strip surface defect detection. Experiments show that this method has higher accuracy and detection efficiency.2. In order to realize detection techniques of the strip surface 3d defects , this paper proposes based on artificial immune system of the strip surface defects detection technology. Based on the detector and antigen domain of space between the containment relationship, and self-body in the domain space of the location information, the introduction of space-based sub-block approach proposed block-based detector generating algorithm, and then defect image three-dimensional gray-scale information to restore the surface to obtain a high degree of information on strip surface defects and thus can achieve the three-dimensional quality of steel strip surface defect detection. Experiments show that the method effectively reduces the detection of loopholes, and eliminate redundant detector and improves the detection efficiency.