Title :
Motion segmentation using pulse-coupled neural network
Author :
Yamaoka, Daisuke ; Ogawa, Yuuki ; Ishimura, Kosei ; Wada, Mitsuo
Author_Institution :
Dept. Chaotic Syst. Eng., Hokkaido Univ., Japan
Abstract :
The pulse-coupled neural network (PCNN) is a neural network which is based on a cat´s visual cortex. We focused on the segmentation ability of the PCNN, and apply the PCNN to the segmentation problem of optical flow. Then, we propose an improved PCNN model with signal generator, and show its performance.
Keywords :
image motion analysis; image segmentation; image sequences; neural nets; signal generators; motion segmentation; optical flow; pulse-coupled neural network; signal generator;
Conference_Titel :
SICE 2003 Annual Conference
Conference_Location :
Fukui, Japan
Print_ISBN :
0-7803-8352-4
