Size measurement of tiny stamping parts based on machine vision
摘要
Aiming at the problem of low efficiency and low accuracy of manual measurement of a kind of micro stamping parts, a measurement method of micro stamping parts based on machine vision is proposed. Firstly, the measurement system design of the system is introduced, and then the measurement method is introduced. The image processing software is used to preprocess the extracted image, such as graying and denoising, and then the Canny edge detection algorithm is used for threshold segmentation to extract the contour of the part. In the contour processing algorithm of these parts, a contour segmentation method based on RDP algorithm is proposed for contour segmentation. In the edge positioning, a method of edge point detection based on caliper tool is proposed to improve the accuracy of edge positioning of various contours. Then the fitting algorithm based on Tukey weight function is used to measure the straight line and arc to obtain the pixel size. Finally, the pixel size is converted to physical size through calibration. Experimental results show that the design has high consistency and accuracy for the measurement of micro stamping parts, and has high efficiency.