Finding and quantitative evaluation of minute bruise on metal surface using hairline

A method to detect minute flaws on metal parts is proposed to remove the defective parts before assembling in a factory. The input gray-scale images of metal parts are used directly to recognize flaws without any image conversion to shorten recognition time. The recognition problem to find flaws and detect their positions on the metal parts is converted here to another problem to search for maximum peak and the variables giving the peak. Then the recognition problem can be treated as an optimization problem, and this conversion allows us to utilize the high performances of Genetic Algorithms in the optimization. The recognition rate of the proposed system was improved by changing the lighting condition so as to emphasize the contrast between the metal surface and the bruise using the reflection character of the hairline on the metal resulted by polishing process. The effectiveness and problems of proposed method are analyzed on standing points of recognition speed and quantitative recognition ability.