Coupled MRF models and CMOS circuits for coarse image-region segmentation

VLSI image/scene recognition systems inspired by brain architecture Combining some processing models inspired by brain vision functions, we have constructed a real-time image/scene recognition VLSI system that can recognize natural scenes including human faces and different objects. The processing functions used in the system are coarse region segmentation using resistive-fuse networks and image recognition of segmented regions using elastic graph matching (EGM). The scene is understood by using the relationship between segmented regions. We have also developed a face and arm gesture recognition system by combining Prof. Miyamoto’s algorithm.