Vision system for part disassembly using a high-speed range sensor

Disassembly of manufactured products has become important in recycling. In this paper, specifications of a vision system for disassembly are studied. A combined system of both range image processing and gray image processing is proposed as vision system for disassembly. An algorithm using extended Gaussian image is proposed to recognize simple features in a range image. Experiments indicate the efficiency of the system using a high-speed range sensor. Fundamental experiments on recycled objects are performed to verify the robustness of the vision system. Bin-picking of cylindrical parts is achieved effectively as an example of disassembly.<
>