A machine vision method for measurement of drill tool wear

As tool wear directly affects the machining quality, tool condition monitoring becomes more and more important in an intelligent manufacturing environment. Chisel edge wear is one of the main wear forms of twist drills. In order to improve the measurement accuracy of chisel edge wear and reduce the cost of detection, this paper proposes a machine vision measurement method for chisel edge wear. A non-local mean denoising method based on integral image and Turky bi-weight kernel function is proposed for the image denoising on the gray distribution of worn tool images. Then the bimodal threshold method and double-threshold Otsu method are proposed to adaptively enhance the image. Finally, the morphological reconstruction-based local extreme point extraction is proposed to effectively complete the tool wear region detection and boundary extraction. The test of drill tool wear in the process of drilling and milling machine is performed to verify the effectiveness of the proposed method. The experimental results show that the proposed method effectively implements the monitoring of tool wear and presents better measurement performance than that of other typical methods.