A Vision-Based Wheel Disc Inspection System

An automated based-vision quality inspection system for wheel-disc is presented in this paper in order to achieve nondestructive and fast defect detection method. The system captures images of wheel-disc which is located on a backlight device. Using these image data, holes in wheel-disc image are detected by ellipse detection algorithm. After that the size and positions of holes on real wheel-disc can be obtained in real time. Experimental results show that although the proposed system is fairly simple, it achieves high accuracy and reliability for wheel disc defect detection.