Passive and active stereo vision for smooth surface detection of deformed plates

Passive stereo vision is proposed to work in a complimentary manner with active stereo vision for detection of smooth surfaces of deformed plates. Passive stereo vision is used to obtain the boundary of the smooth surfaces, whereas active stereo vision with the projection of structured light is applied to detect the details of the surfaces. An inherent problem in passive stereo vision, called false boundary problem, is identified. The problem is solved by calibrating the structured light in active stereo vision and dynamically placing the cameras in passive stereo vision. The matching criteria in active stereo vision and the sensing process of the proposed approach are presented. An experiment was conducted to test the effectiveness of the proposed approach.