A vision-based fusion method for defect detection of milling cutter spiral cutting edge. (June 2021)

• Record Type: Journal Article
• Authors: Zhang, Tongjia; Zhang, Chengrui; Wang, Yanjie; Zou, Xiaofu; Hu, Tianliang
• Is Part Of: Measurement. Volume 177(2021)
• Publication Date: 2021-06
• Subjects: Milling cutter -- Spiral cutting edge -- Defect detection -- Deep learning -- Vision-based; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2021.109248
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: The vision-based fusion method is proposed for defect detection and evaluation of milling cutter spiral cutting edge. Deep learning model is used to extract the target region in order to guarantee the robustness to illumination. Traditional image processing method is used to detect defects and evaluate precision in order to ensure the accuracy. Abstract: Cutting tool is one of the most important parts of machine tool which greatly influences the machining quality. Defect detection of cutting tool spiral cutting edge is usually done by quality control workers after manufactured, which costs time and the quality cannot be guaranteed. In order to detect defects on the spiral cutting edge automatically and reliably within limited cycle time, this paper proposes a vision-based fusion method. This method first uses improved Yolov3-tiny to extract the target cutting edge region, and then traditional image processing method is used to detect and evaluate defects. Compared with only using deep learning, the detection accuracy and evaluation precision of defects are improved. Compared with traditional image processing method, the robustness of illumination is improved. In the case study, the detection result shows that the proposed method can effectively detect and evaluate small defects on the spiral cutting edges illumination insensitively with high detection accuracy.