一种基于奇异值分解的图像匹配算法.pdfVIP

ISSNN 111777TP Journal of Computer Research and Development 7( 1) : 2332, 2010 1, 2 1, 2 3 赵 峰 黄庆明 高 文 1 ( 100190) 2 ( 1000 9) 3 ( 100871) ( fzhao@ jdl. ac. cn) An Image Matching Algorithm Based on Singular Value Decomposition Zhao Feng1, 2 , Huang Qingming1, 2 , and Gao Wen3 1 (Institute of Comp uting Technology, Chinese Academy of Sciences, Beij ing 100190) 2 ( Grad uate University of Chinese A cademy of Sciences, Beij ing 1000 9) 3 (School of lectronics ng ineering and Comp uter Science, Peking Univ ersity , Beij ing 100871) Abstract Image matching plays an important role in many fields, such as computer vision, remote sensing and medical image analysis. Traditionalcorrelationbased matching methods require heavy computation time and they are sensitive to image rotation. In this paper, a new image matching algorithm based on singular value decomposition is proposed to efficiently match image pairs w ith arbitrary rotation angle. Corner points w ith dominant orientation are firstly extracted as feature points in the tw o images. T he initial set of feature point matches is then obtained by singular value decomposition of a correspondence strength matrix. A new expression of the correspondence strength matrix is introduced to handle more complicated imaging conditions. Each element of this matrix is computed from the similarity measure betw een tw o feature points based on normalized cross correlation. T he similarity measure is invariant to image rotation by taking into account the dominant orientation of the tw o feature points. Finally the epipolar geometry constraint is imposed on the initial matches in order to reject the false matches. Experimental results on complicated real images show the effectiveness and robustness of the algorithm. Moreover,