Surface defect recognition using multi-layer perceptron and C-means algorithm

Quality assurance in the powder injection molding is a critical problem due to its complicated processing methods. As surface conditions are major iss ues for the product quality of the powder injection molding, automated visu al inspection on the surface is highly demanded. This paper proposes repres entation and recognition schemes for the surface defects on the powder inje ction molding. From the edge image, line segments were extracted, then they were represented using parameters. Multi-layer perceptron and C-means algo rithm were tested to recognize defective features in the powder injection m olding. The neural network method showed better recognition for the defecti ve features based on the selected measures. Significance: The surface defect in powder injection molding is a critical problem for the product quality assurance. From the complicated surface fea tures, the recognition of defective features were compared between an artif icial neural network and traditional pattern recognition method.