基于识别的手写数字串分割  

Method for handwritten numeral string segmentation based on recognition

作　　者：宋婉娟[1] 董才林[2] 陈增照[2] 何秀玲[2] 张剑[1] 

机构地区：[1]华中师范大学计算机科学系,湖北武汉430079 [2]华中师范大学最优控制与离散数学重点实验室,湖北武汉430079

出　　处：《计算机工程与设计》2007年第21期5198-5200,5217,共4页Computer Engineering and Design

基　　金：湖北省科技攻关计划基金项目(2003BDSP004)。

摘　　要：手写数字串的分割与字符识别密切相关。采用基于识别的分割方法,在分割过程中引入识别机制识别分割碎片,将识别结果经过差值运算后置为每个识别对象的识别可信度,利用动态规划找到最佳分割路径。在训练分类器时,使用反例样本估计分类器参数,得到了性能良好的分类器。实验数据表明,利用正例和反例样本结合训练的分类器比只经过正例样本训练的分类器的识别率要高很多。The segmentation ofhandwritten numeral string is closely linked with character recognition. The recognition-based method is used to solve the segmentation problem of handwritten numeral string. In the segmentation process, recognition mechanism is used to get better results. Recognition confidence is the result of difference operation of outputs of classifier. Based on the confidence of each segmentation candidate combination and dynamic programming searching algorithm, the optimal segmentation combination generates. To get classifier with better performance, negative data must be the necessary trained samples. The experimental results show that the classifier trained with negative data and positive data can get better recognition accuracy than the one trained only with positive data.

关 键 词：数字串分割 反例 动态规划 神经网络 识别可信度 

分 类 号：TP391.43[自动化与计算机技术—计算机应用技术]