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 (22): 232-234. doi: 10.3969/j.issn.1000-3428.2007.22.080

• 人工智能及识别技术 • 上一篇    下一篇

基于USAN的改进的角点检测算法

郭海霞1,解 凯2   

  1. (1. 哈尔滨师范大学数学与计算机学院,哈尔滨 150080;2. 北京印刷学院信息与机电学院,北京 102600)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-11-20 发布日期:2007-11-20

Improved Method for Corner Detection Based on USAN

GUO Hai-xia1, XIE Kai2   

  1. (1. College of Maths and Computer, Harbin Normal University, Harbin 150080; 2. College of Information & Mechanical Engineering, Beijing Institute of Graphic Communication, Beijing 102600)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-20 Published:2007-11-20

摘要: 提出了一种基于USAN的改进的角点检测算法。该算法在原有SUSAN算法的基础上做了如下改进:使用一个3×3的方形预检测窗口对图像的像素进行预检测,在精确检测角点前剔除掉大部分的背景点、边界点及脉冲噪声点,提高了算法的效率;根据图像不同区域对比度不同的特性,采用根据对比度自动调节核心点与其邻域像素的灰度差值门限的方法,使所检测出的角点分布均匀;利用基于USAN定义的角点所应具有的特征(角的边缘及USAN的连续性)来剔除伪角点,降低了角点虚报和漏检的发生率。仿真实验证明了该文所提出的算法具有抗噪能力强、运算量小的特点,适于实时实现。

关键词: 角点检测, USAN, 图像处理, 预检测, SUSAN

Abstract: o avoid the disadvantages of SUSAN, a new algorithm based on USAN is proposed. Corner pre-detection, which utilizes a window of three multiply three for detecting the candidate corners, can reject background, edges and noisy pixels. Intensity discrepancy between nucleus is redefined and its neighbors based on different area have different contrast. This makes the corners well-distributed. False corners are removed according to corner features, which is defined based on USAN. Experimental results show this corner detection method has good capabilities of detection and localization in different contrast image.

Key words: corner detection, USAN, image processing, pre-detection, SUSAN

中图分类号: