摘要
文章提出了一种基于递增权值函数的图像谱的匹配算法,利用递增权值函数,分别对2幅待匹配图像的特征点构造Laplace矩阵,其次进行SVD分解;通过分解后的矩阵特征值和特征向量,寻找匹配矩阵,根据匹配矩阵的特征信息,实现2幅图像特征点之间的匹配;通过对Laplace矩阵和邻接矩阵比较实验,表明了Laplace谱能使发生刚体变换前后的图像获得更高的匹配精度,递增权值函数的Laplace谱比欧式距离的Laplace谱匹配精度要高。
A matching algorithm based on the increasing weighting function of graphic spectrum is proposed in this paper. Firstly, according to the feature points of two related images, the Laplace matrix is defined by the increasing weighting function. Secondly, it is decomposed by SVD. Thirdly, a matching matrix is constructed by the result of the decomposition. Finally, the matching feature points of the two images are obtained according to the matching matrixes. Experiments arc made in order to compare the Laplace matrix with the adjacency matrix. Experiment results indicate that the algorithm in the paper has the higher precision of matching. The results also show that the increasing weighting Laplace spectrum has higher precision of matching than the Euclidean Laplace spectrum.
出处
《合肥工业大学学报(自然科学版)》
CAS
CSCD
北大核心
2008年第11期1778-1781,共4页
Journal of Hefei University of Technology:Natural Science
基金
国家自然科学基金资助项目(10601001
60772121)
安徽省自然科学基金资助项目(050460102
070412065)
安徽省教育厅自然科学研究资助项目(2006KJ030B)