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 ›› 2007, Vol. 33 ›› Issue (16): 166-167,. doi: 10.3969/j.issn.1000-3428.2007.16.058

• 人工智能及识别技术 • 上一篇    下一篇

一种新的基于Hough变换的椭圆轮廓检测方法

周 祥,孔晓东,曾贵华   

  1. (上海交通大学电子工程系,上海 200030)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-08-20 发布日期:2007-08-20

Method of Ellipse Detection Based on Hough Transform

ZHOU Xiang, KONG Xiao-dong, ZENG Gui-hua   

  1. (Electronic Engineering Department, Shanghai Jiaotong University, Shanghai 200030)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-08-20 Published:2007-08-20

摘要: Hough变换是一种检测曲线的有效方法,但使用传统的Hough变换来检测椭圆计算量过大。为此提出了一种新的基于Hough变换的椭圆轮廓检测方法,该方法利用椭圆中心是平面上所有点中距离椭圆轮廓上点最大距离最小的点这一几何性质,降低了Hough参数空间的维度,从而有效地降低了计算量。使用该方法可以快速地判断一个输入图形的轮廓是否为椭圆。

关键词: Hough变换, 椭圆检测, 计算复杂度

Abstract: Hough transform is an effective approach for curve detection. Due to the large computational complexity of ellipse detection using traditional Hough transform, the article presents a method based on Hough transform to decide whether the input figure is an ellipse. The method utilizes a geometric attribute of ellipse, and reduces the statistical dimension of Hough parameter space. As a result, computational complexity decreases and ellipse contour is detected quickly.

Key words: Hough transform, ellipse detection, computational complexity

中图分类号: