Title :
Robust extraction of moving objects through grouping edges along with optical flow
Author :
Yamane, Akihiko ; Ohnishi, Noboru ; Sugie, Noboru
Author_Institution :
Sch. of Eng., Nagoya Univ., Japan
Abstract :
A robust network system is proposed for segmenting and extracting multiple moving objects. The system uses an interconnected neural network in which grouping factors, such as edge proximity and smoothness of edge orientation, and smoothness of motion component perpendicular to an edge, are embedded. Experimental results are provided to indicate performance and noise robustness of the system in extracting objects in synthesized images.
Keywords :
edge detection; image segmentation; image sequences; motion estimation; neural nets; object recognition; edge grouping; edge orientation smoothness; edge proximity; image segmentation; interconnected neural network; motion component smoothness; noise robustness; optical flow; performance robustness; robust moving object extraction; robust network system; Data mining; Humans; Image motion analysis; Image recognition; Image segmentation; Neural networks; Optical computing; Optical fiber networks; Optical noise; Robustness;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.716776
