计算机与现代化 ›› 2013, Vol. 1 ›› Issue (5): 73-76.doi: 10.3969/j.issn.1006-2475.2013.05.017

• 图像处理 • 上一篇    下一篇

基于边缘梯度方向的图像二值化方法

裴沛   

  1. 河海大学计算机与信息学院智能科学与技术研究所,江苏南京211100
  • 收稿日期:2012-12-24 修回日期:1900-01-01 出版日期:2013-05-28 发布日期:2013-05-28

Image Binarization Method Based on Edge Gradient Direction

PEI Pei   

  1. Institute of Intelligence Science and Technology, Computer & Information College, Hohai University, Nanjing 211100, China
  • Received:2012-12-24 Revised:1900-01-01 Online:2013-05-28 Published:2013-05-28

摘要: 针对由于图像背景灰度不均导致不能很好地分离目标与背景问题,在已有工作的基础上,提出一种新的二值化方法。该方法结合Canny边缘检测,基于边缘梯度方向特征信息,通过投票标记边缘两侧和决定不确定点的归属,较好地解决了噪声和伪影问题,实验结果表明该方法具有良好的效果。

关键词: 二值化, 梯度方向, 边缘两侧, 投票

Abstract: As it’s difficult to well separate target from background for images in which background gray level is nonuniform, this paper proposes a new binarization method on the basis of previous work. Combined with Canny edge detection and based on edge gradient direction feature information, the method marks both sides of edge and labels uncertain points by voting to solve the problem of noises and artifacts. Experimental results show that the method works well.

Key words: binarization, gradient direction, both sides of edge, voting

中图分类号: