计算机工程 ›› 2008, Vol. 34 ›› Issue (4): 209-211.doi: 10.3969/j.issn.1000-3428.2008.04.074

• 人工智能及识别技术 • 上一篇    下一篇

一种车牌号码图像二值化的新方法

潘梅森1,张 奋1,雷超阳2   

  1. (1. 湖南文理学院计算机科学与技术系,常德 415000;2. 湖南长沙通信职业技术学院,长沙 410000)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-02-20 发布日期:2008-02-20

New Method for Vehicle License Plate Image Binarization

PAN Mei-sen1, ZHANG Fen1, LEI Chao-yang2   

  1. (1. Department of Computer Science and Technology, Hunan University of Arts and Science, Changde 4150002. Changsha Tel. & Tec. Vocational College, Changsha 410000)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-02-20 Published:2008-02-20

摘要: 利用数学形态学的开运算消除图像光照不均匀影响,根据欧几里德距离确定图像阈值。基于人类视觉特性确定每个像素点的视觉敏感度系数,并根据图像像素点灰度值、图像阈值和视觉敏感度系数对图像进行二值化处理。该方法能有效地消除伪影,较好地保护字符笔画。模拟实验结果表明,该方法是有效的。

关键词: 二值化, 阈值, 人类视觉系统, 像素

Abstract: This paper eliminates the image non-uniform brightness using mathematical morphological open operation and obtains the image threshold value according to the Euclidean distance. Standard deviation is used to determine the visual sensitivity coefficient of each pixel according to the human visual characteristics. Image binarization is performed according to the gray value, the image threshold and the visual sensitivity coefficient. This method both can effectively eliminate the ghosts, and can keep the character stroke well. Experimental results show that this method is more effective.

Key words: binarization, threshold, human vision system, pixel

中图分类号: