利用改进的SUSAN算法提取航空影像中孤立的特征点Extracting isolated feature points in aerial images by using improved SUSAN algorithm
杜艺,龚循平
摘要(Abstract)：
本文针对地形起伏较大、无明显建筑物的航空影像,分析了SUSAN算法角点检测理论,提出一种提取孤立特征点的方法。该方法先对图像进行梯度幅值运算,然后对梯度幅值进行Otsu法阈值分割,设计模板并对孤立特征点进行套合,最后利用SUSAN算法计算原始影像的角点初始响应,经过非极大值抑制提取孤立特征点。经实验证明,与传统的Harris角点、Forstner角点相比,该特征点受地形起伏、太阳高度角、视角变换等外界条件干扰较小,为下一步影像匹配做了较好的准备。
关键词(KeyWords)： 特征点提取;SUSAN算法;角点检测;阈值分割
基金项目(Foundation):
作者(Author): 杜艺,龚循平
DOI: 10.16251/j.cnki.1009-2307.2011.06.015
参考文献(References)：
- [1]Rodehorst V,Koschan A.Comparison and evaluation offeature point detectors[C]//Proc.5th InternationalSymposium Turkish-German Joint Geodetic Days“Geod-esy and Geoinformation in the Service of our DailyLife”,Technical University of Berlin,ISBN 3-9809030-4-4,Berlin,Germany.,2006.
- [2]Moravec H P.Toward automatic visual obstacle avoidance[C]//Proc.5th Int.Joint Conf.Artificial Intelligence,Cambridge,USA,1977:584.
- [3]Schmid C,Mohr R and Bauckhage C.Evaluation ofinterest point detectors[J].Int.Journal of Computer Vision,2000,37(02):151-172.
- [4]Harris C and Stephens M.A combined corner and edgedetector[C]//Proc.4th Alvey Vision Conf.1988:147-151.
- [5]M Trajkovi,M Hedley.Fast corner detection[J].Im-age and Vision Computing,1998,16(2):75-87.
- [6]Smith S M and Brady J M.SUSAN-a new approach to lowlevel image processing[J].Int.Journal of ComputerVision,1997,23(01):45-78.
- [7]冈萨雷斯.数字图像处理[M].北京:电子工业出版社,2008:105-106.