• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2008, Vol. 30 ›› Issue (1): 66-68.

• 论文 • 上一篇    下一篇

人民币号码自动识别方法研究

段敬红 栾丹   

  • 出版日期:2008-01-01 发布日期:2010-05-19

  • Online:2008-01-01 Published:2010-05-19

摘要:

针对人民币印刷中号码的识别问题,本文研究了其中的识别算法,包括号码图像的二值化、号码区域的定位、号码字符的分割及归一化和号码字符的识别,提出了一种号码区域定位及分割方法。在此基础上,建立了相应的识别系统并进行了实验研究。结果表明,系统具有较高的识别精度,能较好地满足实际需求。

关键词: 号码识别 图像分割 特征提取

Abstract:

The number recognition method for RMB notes in the press process has been studied. The recognition algorithm includes image binarization, number region localization, number characters' extraction and normalization, number feature extraction and the recognition. A method for number region localizationn and segmentation is proposed. Based on the algorithm, a recognition system is built and the relevant experiments is investigated. The experimental results show the system has a high recognition rate.

Key words: number recognition, image segmentation, feature extraction