A Study on Identification of Bubble Movements in an Automatic Wheel Leakage Detection System

This paper describes a method to detect leaks on the surface of a wheel rim manufactured by aluminum die casting and forging. Leaks in the wheel rim may be caused by temperature gradient differences during the cooling period, impurities, or flowing air bubbles in the die casting process. We developed a new low-cost approach to detect leaks that only required a short inspection time by observing the movement of soap bubbles on the wheel rim surface. We designed new light sources to minimize mirror image effects and irregular luminance distributions on the curved wheel rim surface. A preprocessing procedure for image processing was also developed to determine the bubble configurations using threshold and morphology techniques. A series of experiments were conducted to evaluate the performance of the inspection system. The results demonstrated that leaks generated during the wheel rim manufacturing process could be accurately detected even under various environmental lighting conditions.