In-Process Tool Wear Measurement System Based on Image Analysis for CNC Drilling Machines

Abstract

Tool condition monitoring (TCM) has been a constant field of research. Conventionally, some sensors are installed at specific parts of the machine, and by using the signal-processing techniques, the tool wear is estimated. In this article, a direct system based on image analysis has been developed to automate the in-process tool wear measurement. The method uses only a single camera installed inside the machine and a tree-stage measurement process composed of image treatment, image comparison, and wear measurement. Experimental results show that the detection of similar images has a success index rate (SIR) equal to 98.89%, whereas the measurement error of the average flank wear and the maximum flank wear is estimated to be 3.57% and 2.92%, respectively.