作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2006, Vol. 32 ›› Issue (6): 203-205,208.

• 人工智能及识别技术 • 上一篇    下一篇

改进的多阈值动态二值化算法

马 驰 1,张红云2,苗夺谦2,张学东1   

  1. 1. 鞍山科技大学计算机科学与工程学院,鞍山 114044;2. 同济大学计算机科学与工程系,上海 200092
  • 出版日期:2006-03-20 发布日期:2006-03-20

Improved Multi-threshold and Dynamic Binarization Algorithm

MA Chi1, ZHANG Hongyun2, MIAO Duoqian2, ZHANG Xuedong1   

  1. 1. School of Computer Science and Engineering, Anshan Science and Technology University, Anshan 114044;2. Department of Computer Science and Engineering, Tongji University, Shanghai 200092
  • Online:2006-03-20 Published:2006-03-20

摘要: 在模式识别中,二值化效果的好坏直接影响着识别率。文章在研究现有二值化算法的基础上,对多阈值动态二值化进行了改进,并以身份证为例来验证该算法的有效性,实验结果表明所提算法有较高的抗噪音能力。

关键词: 多阈值;动态;二值化;身份证

Abstract: In pattern recognition, the effect of binarization directly affects the recognition rate. In the foundation of binarization algorithms research, the paper improves the multi-threshold and dynamic binarization algorithm. Taking example for ID card, it validates the effect of this algorithm. Experimental result indicates that the proposed method has better capability of anti-noise.

Key words: Multi-threshold; Dynamic; Binarization; ID card