Abstract:  When computer vision technology was applied to high-speed steel(HSS)saw tooth edge detection, it was impossible to rely on the sawtooth edge to detect the sawtooth parameters. For this reason, an automatic detection device based on image processing and a detection algorithm to improve its detection accuracy were proposed. Firstly, the hardware structure of the detection device was designed, which was composed of five parts: sawtooth shooting, saw blade clamping, saw blade rotation, saw blade horizontal detection and saw blade horizontal adjustment. Then the software of the detection device was designed, which was composed of three steps: image preprocessing, sawtooth edge detection and sawtooth parameter calculation. To solve the problem of large identification error of sawtooth parameters caused by gray value noise on the surface of highspeed steel sawtooth, the maximum inter class and intra class variance ratio method were introduced to adaptively adjust the optimal gray threshold, so the saw tooth edge could be accurately detected, and the saw tooth parameters could be accurately calculated. Finally, a sawtooth parameter measurement device was built, and the feasibility of the sawtooth parameter measurement device and the correctness of the detection algorithm were verified through experimental methods. The results show that the detection accuracy of the detection device proposed was 3%, which is close to the detection results of other equipment such as three coordinates. Therefore, the proposed detection device can be used as a basic tool for the application of high-speed steel circular saw blade selection, saw blade grinding and detection, and so on.

Key words: high-speed steel(HSS) circular saw blade;saw tooth parameters;edge detection; image processing

LI Qin, LU Di-fei, CHEN Gui-qiang, et al. Automatic detection device and algorithm for saw tooth parameters of HSS circular saw blade［J］.Journal of Mechanical & Electrical Engineering, 2022,39(1):107-111.