Title :
A handwritten character recognition method with ANN feedback
Author :
Xiaoyan, Zhu ; Yifan, Shi
Author_Institution :
Dept. of Comput. Sci. & Technol., Tsinghua Univ., Beijing, China
fDate :
6/23/1905 12:00:00 AM
Abstract :
In this paper, a new approach of handwritten character recognition system with Artificial Neural Network (ANN) feedback is proposed. This recognition system is based on a neural recognition network and a neural feedback network, the out put of which is used to modify the feature of the input pattern, thus preprocessing and recognition are integrated closely. Experiments show that this approach can make system performance very good and robust to environmental noise
Keywords :
feedback; handwritten character recognition; neural nets; optical character recognition; artificial neural network feedback; handwritten character recognition method; neural feedback network; neural recognition network; Artificial neural networks; Character recognition; Control systems; Feature extraction; Feedback loop; Neurofeedback; Noise robustness; Pattern recognition; Signal processing; System performance;
Conference_Titel :
Document Analysis and Recognition, 2001. Proceedings. Sixth International Conference on
Conference_Location :
Seattle, WA
Print_ISBN :
0-7695-1263-1
DOI :
10.1109/ICDAR.2001.953794
