摘要
针对处理重复内容、非单一平面场景时传统方法常见的误匹配、漏匹配问题,提出基于特征点拓扑结构及三角剖分的无约束场景特征匹配方法.利用相似内容比非相似内容具有明显更近特征描述欧氏距离的特点,提出K近邻距离比算法,保留具有明显更小灰度及梯度差异的多对多特征点对作为初匹配,以减少漏匹配;通过特征点三角剖分的映射去除K近邻距离比初匹配中一对一误匹配;根据两图特点集的拓扑相似性度设计基于拓扑的分级三角剖分算法,并对K近邻距离比初匹配中多对多匹配进行一对一确认,得到无约束场景特征匹配结果.实验结果表明,该方法可同时显著抑制误匹配和漏匹配.
Repeated contents and non-single-plane scenes usually lead to serious mismatching or missed matching. A novel feature matching method using topology structure and triangulation is proposed to match images without any constraint conditions. First, using Euclidean distance to measure similarity, similar features are significantly closer than dissimilar contents. K nearest neighbors distance ratio algorithm is proposed to find matching candidates which have significantly smaller distances. Second, 1-to-1-matches in reference image are triangulated, and the triangulation is mapping to the target image. The outliers which do not conform to the triangulation rules are moved. Then, all feature-sets are hierarchically triangulated based on topology similarity measure, and m-to- n-matches are reduced to 1-to-1-matches. The experimental results show that with the proposed method, mismatching and missed matching can be greatly reduced.
引文
[1]Mikolajczyk K,Schmid C.Scale affine invariant interest point detectors[J].International Journal of Computer Vision,2004,60(1):63-86
[2]Belongie S,Malik J,Puzicha J.Shape matching and object recognition using shape contexts[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,2(4):509-522
[3]Ke Y,Sukthankar R.PCA-SIFT:a more distinctive representation for local image descriptors[C]//Proceedings of IEEE Computer Society,Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE computer Society Press,2004,2:506-513
[4]Lazebnik S,Schmid C,Ponce J.A sparse texture representation using affine-invariant regions[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE computer Society Press,2003,2:319-324
[5]Lowe D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91–110
[6]Schaffalitzky F,Zisserman A.Multi-view matching for unordered image sets,or“how do I organize my holiday snaps?”[M]//Lecture Notes in Computer Science.Heidelberg:Springer,2002,2350:414-431
[7]Gool V L,Moons T,Ungureanu D.Affine/photometric invariants for planar intensity patterns[M]//Lecture Notes in Computer Science.Heidelberg:Springer,1996,1064:642-651
[8]Mikolajczyk K,Schmid C.A performance evaluation of local descriptors[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(10):1615-1630
[9]Tzimiropoulos G,Argyriou V,Zafeiriou S,et al.Robust FFT-based scale-invariant image registration with image gradients[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(10):1899-1906
[10]Yu G S,Morel J M.A SIFT:An algorithm for fully affine invariant comparison[OL].2014-03-04.http://dx.doi.org/10.5201/ipol.2011.my-asift
[11]Zhai Y,Zeng L.A SIFT matching algorithm based on adaptive contrast threshold[C]//Proceedings of International Conference on consumer Electronics Communications and Networks.Los Alamitos:IEEE Computer Society Press,2011,1934-1937
[12]Wei Hu,Zhang Liyan,Liu Shenglan,et al.An algorithm on registration of multi-view range images based on SIFT feature matching[J].Journal of Computer-Aided Design&Computer Graphics,2010,22(4):654-661(in Chinese)(韦虎,张丽艳,刘胜兰,等.基于SIFT图像特征匹配的多视角深度图配准算法[J].计算机辅助设计与图形学学报,2010,22(4):654-661)
[13]Beis J S,Lowe D G.Shape indexing using approximate nearest-neighbor search in high-dimensional spaces[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,1997:1000-1006
[14]Silpa A C,Hartley R.Optimised k D-trees for fast image descriptor matching[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2008:1-8
[15]Yang S W,Wang C C,Chang C H.RANSAC matching:simultaneous registration and segmentation[C]//Proceedings of the IEEE International Conference on Robotics and Automation.Los Alamitos:IEEE computer Society Press,2010:1905-1912
[16]Ji Hua,Wu Yuanhao,Sun Honghai,et al.SIFT feature matching algorithm with global information[J].Optics and Precision Engineering,2009,17(2):439-444(纪华,吴元昊,孙宏海,等.结合全局信息的SIFT特征匹配算法[J].光学精密工程,2009,17(2):439-444)
[17]Jia J Y,Tang C K.Image stitching using structure deformation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(4):617-631
[18]Ma J,Ahuja N.Region correspondence by global configuration matching and progressive Delaunay triangulation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2000,2:637-642
[19]Li Y,Tsin Y,Genc Y,et al.Object detection using 2D spatial ordering constraints[C]//Proceedings of IEEE Computer Society,Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE computer Society Press,2005,2:711-718
[20]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
