Backlit mouse defect inspection using machine vision

In this paper, we developed a robust machine vision system for backlit mouse defect inspection. The defects might include incorrect illuminating area, incorrect LED color saturation, fragment or crack lighting pattern on the backlit mouse. A set of machine vision algorithms were designed to segment the inspection regions (IRs) and extract three features of the segmented IRs. These features were area of illuminating region, mean red value, and shape similarity. Two modes based machine vision system, engineer mode for testing-phase and operator mode for inspection-phase, were used to inspect the mentioned defect. Some experiments were given to show the efficiency and effectiveness of the proposed backlit mouse defect vision inspection system.