摘要
为了提高图像匹配方法的效果,提出了一种面向图匹配的属性关系图模型,该模型利用特征点集的分布情况构建特征点和边的属性关系.首先用特征点与点集中心点连线一侧的特征点数目作为特征点的属性,再根据特征点间连线两侧的点数目大小来指定边的方向关系和属性信息;然后构造两幅图像之间的亲和矩阵;最后利用整数约束下的迭代求解方法求解匹配结果.实验结果表明该算法具有很好的性能,在形变大的图像上也有很好的匹配效果.
In order to improve the effect of image matching method,an attribute relational graph model for graph matching is proposed. The graph model uses the distribution of feature point set to construct the attributed relationship between feature points and edges. Firstly,the number of feature points on the side of the feature point and the center point of the point set is used as the attribute value of the feature point. Then the direction relationship and attribute information of the edge are specified according to the number of points on both sides of the line connecting the feature points,and constructe the affinity matrix between the tw o images; Finally,the iterative solution method under integer constraints is used to solve the matching results. The experimental results showthat the algorithm has good performance and has a good effect on images with large deformation.
引文
[1]Qin Xu-jia,Hong Xia-yang,Wang Hui-ling,et al.Improved SURFfeature matching algorithm for remote sensing images[J].Journal of Chinese Computer Systems,2016,37(2):327-331.
[2]Miao Quan,Cheng Guang.Image matching algorithm based on local feature descriptor using gray level histogram[J].Journal of Chinese Computer Systems,2016,37(7):1567-1571.
[3]Conte D,Foggia P,Sansone C,et al.Thirty years of graph matching in pattern recognition[J].International Journal of Pattern Recognition and Artificial Intelligence,2004,18(3):265-298.
[4]Singh K,Singh V.Graph pattern matching:a brief survey of challenges and research directions[C]//2016 3rd International Conference on Computing for Sustainable Global Development,New Delhi,INDIA,2016:199-204.
[5]Mayvan B B,Rasoolzadegan A.Design pattern detection based on the graph theory[J].Knowledge-Based Systems,2017,120:211-225.
[6]Tian Y,Yan J,Zhang H,et al.On the convergence of graph matching:graduated assignment revisited[J].European Conference on Computer Vision,2012,7574(1):821-835.
[7]Jiang B,Tang J,Cao X,et al.Lagrangian relaxation graph matching[J].Pattern Recognition,2016,61:255-265.
[8]Feng W,Liu Z Q,Wan L,et al.A spectral-multiplicity-tolerant approach to robust graph matching[J].Pattern Recognition,2013,46(10):2819-2829.
[9]Leordeanu M,Hebert M.A spectral technique for correspondence problems using pairw ise constraints[C]//10th IEEE International Conference on Compute Vision,2005,2:1482-1489.
[10]Aguilar W,Frauel Y,Escolano F,et al.A robust graph transformation matching for non-rigid registration[J].Image and Vision Computing,2009,27(7):897-910.
[11]Van WYK BJ,Van WYK M A.A POCS-based matching algorithm[J].IEEE Transaction on Pattern Analysis and Machine Intelligence,2004,26(11):1526-1530.
[12]Zhu Ming,Liang Dong,Tang Jun,et al.Point pattern matching algorithm based on Q-spectrum of line graph[J].Journal of South China University of Technology,2011,39(7):102-108.
[13]Xu Yang,Hong Qiao,Liu Zhi-yong.Feature correspon-dence based on directed structural model matching[J].Image and Vision Computing,2015,33(1):57-67.
[14]Feng Zhou,Fernando De la Torre.Factorized graph matching[J].IEEE Transactions on Pattern Analysis and M achine Intelligence,2016,38(9):1774-1789.
[15]Leordeanu M,Hebert M,Sukthankar R.An integer projected fixed point method for graph matching and M AP inference[C]//Neural Information Processing Systems,2009:1114-1122.
[16]Leordeanu M,Hebert M.A spectral techniquefor correspondence problems using pairw ise constraints[C]//International Conference on Computer Vision(ICCV),2005:1482-1489.
[17]Zass R,Shashua A.Probabilistic graph and hypergraph matching[C]//IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2008:1-8.
[1]秦绪佳,洪夏阳,王慧玲,等.改进的遥感图像SURF特征匹配算法[J].小型微型计算机系统,2016,37(2):327-331.
[2]苗权,程光.面向图像匹配的局部灰度直方图特征描述子[J].小型微型计算机系统,2016,37(7):1567-1571.
[12]朱明,梁栋,唐俊,等.基于线图Q-谱的点模式匹配算法[J].华南理工大学学报,2011,39(7):102-108.