论文部分内容阅读
方向线素特征在手写体汉字识别中得到了广泛的应用,文中对该特征进行了形式化描述,分析了其存在的不足,在普通方向线素特征的基础上,引入模糊数学思想,提出了汉字的模糊方向线素特征,给出了形式化描述,应用该特征实现了一个非特定人手写体汉字识别系统,经对含有37550个汉字(10套国标一级汉字)的未学习样张的测试,识别率由85.5%提高到89.7%,十选正确率达到98.4%,在同等测试条件下,比单纯使用方向线素特征识别率提高了四个百分点
The characteristics of the directional algebra have been widely used in the recognition of handwritten Chinese characters. In this paper, the characteristics are formally described and their deficiencies are analyzed. Based on the features of normal directional algebra, fuzzy mathematics is introduced and the characters of Chinese characters , This paper gives a formal description of the fuzzy directional linear features. By using this feature, a non-specific handwritten Chinese character recognition system is implemented. After testing and identifying the unlearning proofs containing 37,550 Chinese characters (10 sets of GB first-class Chinese characters) Rate increased from 85.5% to 89.7%, the ten-election correct rate of 98.4%, under the same test conditions, than simply using the direction of the linear feature recognition rate increased by four percentage points