作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

›› 2004, Vol. 30 ›› Issue (02): 70-72.

• 论文 • 上一篇    下一篇

一种基于全局均值和局部方差的图像二值化方法

龙钧宇,金连文   

  1. 华南理工大学电子与信息学院,华南理工大学电子与信息学院 广州 510640
  • 发布日期:2004-01-25

An Image Binarization Method Based on Global Mean and Local Standard Deviation

LONG Junyu,JIN Lianwen   

  1. (College of Electronics and Information, South China University of Technology, Guangzhou 510640)
  • Published:2004-01-25

摘要: 比较了几种常用的动态阈值二值化方法,并在此基础上提出了一种新的基于均值和方差的二值化方法。试验表明,这种方法能够较好地区分物体和背景,并且相对于原来的逐点进行二值化的方法而言大大提高了处理速度。

关键词: 二值化方法, 数学期望, 方差

Abstract: In this paper, several widely used binarization methods are studied and compared, and a new binarization method that is based on mean and standard deviation of a given image is proposed. Experiments show that this method can distinguish the object and the background perfectly, and being compared to the method that calculates a threshold for each pixel and binarizes the pixel one by one, the proposed method can improve the processing speed greatly.

中图分类号: