| The process of tool wear dulls a tool cutting edge, increases friction between the tool and the workpiece, and increases power consumption. This thesis presents an effective machine vision system for monitoring wear and detecting breakage in single point cutting tools (CNC lathe tools). The newly developed algorithm is implemented in Microsoft C V5.1 and the system is composed of a microscope, an optic fiber illuminator, a Zenith IBM/XT compatible micro-computer with an Image Capture Board, a CCD RGB video camera and a colour monitor.; The image of a given tool is converted into a binary image accentuating the wear land. Then, five features are extracted and used in a classification scheme. A microcomputer reports on the state of the tool based on the results of the classification. The detection of tool breakage and the ease of integration with existing monitoring techniques are some of the advantages of the developed system. (Abstract shortened by UMI.)... |
