Title :
A neural network approach to the consistent labeling problem
Author :
Feng, Jiansheng ; Tao, Wenjing ; Burkhardt, Hans
Author_Institution :
Tech. Inf. I, Tech. Univ. of Hamburg, Germany
Abstract :
A method is presented to solve the consistent labeling problem, especially the high-order constraint case, by the Hopfield network. An agitating method is proposed to address the problem of local minima. Some experiments concerning line drawing labeling are demonstrated
Keywords :
Hopfield neural nets; image recognition; set theory; Hopfield network; agitating method; consistent labeling; line drawing labeling; local minima; neural network; Artificial intelligence; Capacitive sensors; Computer architecture; Educational institutions; Hopfield neural networks; Labeling; Layout; Neural networks; Neurons; Remote sensing;
Conference_Titel :
Neural Networks, 1993., IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0999-5
DOI :
10.1109/ICNN.1993.298636
