Title :
Local feature-based recognition of partially occluded objects using neural network
Author :
Zheng, Naming ; Li, Yaoyong ; Houwers, Wiek P M
Author_Institution :
Inst. of AI & Robotics, Xi´´an Jiaotong Univ., China
Abstract :
A new method of recognizing partially occluded objects using neural networks is presented. The neural network consists of a simplified ART-2 and a two-layer feedforward network, and its inputs are the local features of objects. The network is first trained using a set of local features of known objects, then it can be used to recognize unknown object(s). Our numerical experiments using this method show encouraging results, especially for recognizing the occluded objects
Keywords :
ART neural nets; feature extraction; feedforward neural nets; learning (artificial intelligence); object recognition; ART-2 network; local feature-based recognition; neural network; object recognition; partially occluded objects; training; two-layer feedforward network; Artificial neural networks; Cameras; Computer vision; Feature extraction; Flexible manufacturing systems; Humans; Image edge detection; Layout; Machine vision; Neural networks;
Conference_Titel :
Industrial Electronics, Control, and Instrumentation, 1995., Proceedings of the 1995 IEEE IECON 21st International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-3026-9
DOI :
10.1109/IECON.1995.483985