An improved photometric stereo method for the online defect detection of gearbox covers

The defects on the surface of a gearbox cover can affect its high performance and high air-tightness. To ensure its high functionality, it's necessary to detect the surface defects. The photometric stereo (PS) method is widely applied to defect detection in the industry for its high accuracy and high speed. However, the traditional PS simply applies to static scenes. To solve this problem, we proposed an improved photometric stereo method that can detect defects in dynamic scenes with high speed and high accuracy. Experiments are designed to verify the feasibility of our method and the results showed that our method performs well with objects with complex shapes.