An image inspection method for car brake parts

In order to increase productivity and efficiency, the components factory for car manufacture needs an automatic flaw detecting system by using image processing. The conventional method could inspect flaw of a metal surface with the tiny ruggedness and the dirt which remained in the washing process. The result of a long-term test proved parts with which the inside of a flaw is dark because of decline in the reflectivity of the surface where time has passed or accumulation of dirt existed, and the conventional method was difficult to detect the flaw of such parts. This study proposes the automatic flaw detecting method which can detect those flaws by adding a new procedure to the process which used the rate of the luminance of the conventional method. By the experimental results, it was confirmed that the flaws of defective parts used as the inspecting standard can be extracted certainly.