基于遗传算法的手写体汉字识别系统优化方法的研究

被引:13
作者
林磊
王晓龙
刘家锋
机构
[1] 哈尔滨工业大学计算机科学与工程系!哈尔滨
关键词
手写体汉字识别; 遗传算法; 参数优化;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
为了改善手写体汉字识别系统的性能 ,提出了前端单字识别器 (SCR)和后端语言解码器 (post- processingsystem )有效结合的模型 ,并且利用遗传算法对系统参数进行优化 .以联机手写体汉字识别系统作为 SCR进行测试 ,首选准确率为 6 9.46 % ,汉字识别的准确率达到 87.5 9% ,较优化前提高 6 .4% .实验结果表明 ,遗传算法 (GA )是一种有效的优化系统参数的方法 .
引用
收藏
页码:658 / 661
页数:4
相关论文
共 9 条
  • [1] The post -processing statistical method of Chinese text recognition. Ying Xia. Pattern Recognition . 1996
  • [2] The system of online handwritten Chinese character recognition based on association. Tang Xianglong,Shu Wenhao,Liu Jiafeng et al. Information and Control . 1991
  • [3] Adaptation in Nature and Artificial Systems. Holland JH. . 1975
  • [4] Methods of combining multiple classifiers and their applications to handwriting recognition. Xu Lei,Krzyzak Adam,Suen Ching Y. IEEE Transactions on Systems Man and Cybernetics . 1992
  • [5] Pattern recognition.[P]. DIEP THANH A;AVI-ITZHAK HADAR I;GARLAND HARRY T.EP0622750A2,1994-11-02
  • [6] A post -processing system by use of context information.In: Proc of Int’ l Conf on Chinese Information Processing. Xu Zhoujun,Fang Yingqian. Beijing Review . 1998
  • [7] Analyzing and improving statistical language models for speech recognition[Ph D dissertation]. Ueberla J. . 1994
  • [8] Estim ation of probabilities from sparse data for thelanguage model component of a speech recognizer. Katz S M. IEEE Transon Acoustics,Speech and Signal Processing . 1995
  • [9] Genetic algorithms. Holland JH. The Scientific American . 1992