Dynamic Balance Detection Method of High-speed Cutting Tool Based on Computer Vision

With the increasing demand for high-speed cutting technology in modern manufacturing industry, tool dynamic balance has become a key factor affecting cutting efficiency, machining quality and tool life. In this paper, a dynamic balance detection method of high-speed cutting tools based on computer vision is proposed to overcome the limitations of traditional dynamic balance detection methods and improve the detection accuracy and efficiency. This method achieves rapid acquisition and high-quality processing of tool images by constructing a computer vision system consisting of Panasonic LUMIX series high-resolution industrial cameras, Sigma high-definition industrial lenses, LR-LINK quad-port 5G PoE image acquisition cards, and dedicated LED light sources. In the research process, Gaussian filter and Canny edge detection algorithm are used to preprocess the image to improve the accuracy of feature extraction. Through ellipse fitting and centroid calculation, the unbalance of the tool is accurately obtained. Finally, according to the calculation results, suggestions for dynamic balance adjustment are given. In the experiment, three different types of tools, namely end milling cutter, drill bit and turning tool, were selected and verified under various cutting conditions. The results show that the detection error rate of this method is reduced by about $\mathbf{1 5 \%}$ to $\mathbf{2 5 \%}$ compared with the traditional dynamic balance detection method, and it has higher accuracy and stability. In addition, computer vision technology realizes automatic detection, avoids potential damage to the tool during the detection process, and significantly improves the detection efficiency. This study provides an efficient and accurate new method for dynamic balance detection of high-speed cutting tools, which has important application value.