Vision-based bin-picking: recognition and localization of multiple complex objects using simple visual cues

摘要

Vision-based bin-picking is increasingly more difficult as the complexity of target objects increases. We propose an efficient solution where complex objects are sufficiently represented by simple features/cues; thus invariance to object complexity is established. The region extraction algorithm utilized in our approach is capable of providing the focus of attention to the simple cues as a trigger toward recognition and pose estimation. Successful bin-picking experiments of industrial objects using stereo vision tools are presented.