Research On Bore Surface Inspection Based On Pattern Recognition

Chamber of Cannon surface detection is an important indicator of the artillery test items. The bore surface defects is one of the key factors that affect the quality and performance of the barrel. In order to improve the current cannon barrel bore defect detection. For example still recognition by the human eye chamber body defects, and other means of detection is complex operation. The paper design based on digital image processing technology and pattern recognition technology, CCD imaging technology, the use of the multidisciplinary intervention software platform photoelectric bore detection system.The article pretreatment of the image by image processing technology. The decision tree classifier in pattern recognition the completed feature extraction. Defect identification and analysis of rust, and its defect classification. Thus achieving the purpose of quantitative detection of the bore surface defect detection.Bore surface defects as a research subject, using image processing and pattern recognition algorithms. Calculate and identify the bore surface defects, and establishment of inner bore surface detection system. Through the experimental verification of the inner bore detection system based on pattern recognition is feasible.Combined with pattern recognition technology and image processing technology development based on bore crawling on the bore surface defect detection system software. The system completes detection experiments related to C++and HALCON software platform. The implementation results show that changing the system is able to precisely identify the defects such as the bore surface scratches, abrasions, rust and corrosion. And the system identification of the defect rate is as high as98%. The system has simple operation, fast speed, has high value for engineering.