一种新的票据自动识别系统

A Novel Automatic Recognition System of Bills

设计了一种基于主曲线和粗糙集方法相结合的新颖的票据自动识别系统.首先提出一种高效的数字框定位技术、边框处理方法,并用于从票据中提取待识别数字串;然后建立有效的数字串分割算法来分割数字串;最后利用主曲线和知识约简算法进行数字特征提取和分类器设计.实验结果表明,该方法在实践中是可行的,为各种票据自动识别提供了一条新途径.

A new automatic recognition system of bills is designed on the basis of the principal curves and rough set methods. Firstly, we propose a high-efficient technique by which the location of the digit frame can be accurately detected and an irrelevant streaks removal method. They can be used to extract recognized digit string from the bill; Secondly, an efficient digit string segmentation algorithm is set up to segment the digit string; Finally, The algorithms of principal curves and knowledge are used to extract the structural features of digits and design a classifier. Experimental result indicates that the method is feasible in practice, and provides a new approach to the automatic recognition of various kinds of bills.