基于改进Canny算法的转辙机缺口检测系统研究OA

In view of the large workload of manual detection of railway switch machine notches and the fact that measurement accuracy is easily affected by human factors,a time detection system for switch machine notches is proposed based on an improved Canny algorithm.The system hardware consists of an industrial control computer,a vision camera,and a background light source.The software is developed based on an improved Canny edge detection algorithm combined with the algorithm platform of Hikvision VisionMaster.The software design uses adaptive median filtering to remove noise interference,extracts image edge features through pixel gradient values and non-maximum suppression,and uses the maximum inter-class method to adaptively select the high and low thresholds of the switch machine notch image.The VisionMaster algorithm platform is used for image calibration,and the edge contour of the notch image is measured line by line,and then converted into the actual physical size of the switch machine notch through calibration.Experimental tests show that the switch machine notch detection system using the improved Canny algorithm can effectively identify the edge contour of the switch machine notch image and measure the size of the notch in real time.