本文探讨一种新的在工程图扫描图象中自动识别字符、符号的方法,该研究是基于点相关的神经网络识别技术,这种方法考虑了各种退化的字符样本,训练一个字符时使用一种新的学习规则来自动完成训练过程,并从一组字符图象样本集中产生每个字符的理想的特征描述。这种方法使学习的复杂度呈常量,并在工程图字符识别中得到实际的应用。
Abstract
This paper explores a new method for automatic recognixing characters and symbols in the engineering drawing scanning image which is based on the point correlation neural network technology. This method has taken into account the degenerated character samples ,and attempts to complete automatically the training process by a new learning rule ,to create an optimal feature representation of each character from a set of character sample image. The research makes the learning complexity constant ,and gets effective application in the engineering drawing character recognition.
关键词
图纸扫描 /
图象分割 /
神经网络 /
字符识别
{{custom_keyword}} /
Key words
Drawings scanning /
image segmentation neural network /
character recognition
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Wilson S S. Theory of metrix morphology , IEEE Transactions on pattern Analysis and Machine Intelligence ,Sept. 1992
[2] Wilson S S. Teaching network connectivity using simulated annealing on a massively parallel processor. Proceeding IEEE ,1991 ,79 (4) :559 - 566
[3] Haralick R M et al . Image amalysis using mathematical morphology. IEEE Transactions on Pattern Analysis and Machine Intelligence ,1988 ,9 (4) :532 - 550
[4] Thomas L Hemminger et al . Using a hopfild network for rotation and scale independent pattern recognition. Neural Network International Conference ,1994 ,3003 - 3008
[5] Yoshimasa K et al . A new scheme which imcrementally generates networks for diwtorted handprinted kanji pattern recognition. Neural Network International Conference ,1994 ,3852 - 3855
[6] Lalit gupta. A Neural network approach to robust shape classification. IEEE Pattern Recognition ,1990 ,23 (6) :455 - 465
[7] Etienne Barnard. Image processing for image understanding with neural nets. IEEE Neural Network International Conference ,1990, 202 - 212
[8] Chan Pyng Lai et al . Recognition of dimension sets in engineering drawings. IEEE Transaction on Pattern Analysis and Machine Intelligence ,1994 ,16 (8) :848 - 855
[9] Jairo Rocha et al . Character recognition without segmentation. IEEE Transaction on Pattern Analysis and Machine Intelligence ,1995 ,17 (9) :903 - 909
[10] Nagasamy Vijay. Engineering drawing processing and vectorization system. Computer Vision , Graphics & Image Processing ,1990 ,49 :379 - 397.
[11] 焦李成. 神经网络的应用与实现. 西安:西安电子科技大学出版社,1995
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(426 KB)
