基于加权相似性度量的特征匹配方法

doi:10.11772/j.issn.1001-9081.2020050747

计算机应用 ›› 2021, Vol. 41 ›› Issue (2): 511-516.DOI: 10.11772/j.issn.1001-9081.2020050747

所属专题：多媒体计算与计算机仿真

• 多媒体计算与计算机仿真 • 上一篇下一篇

基于加权相似性度量的特征匹配方法

胡立华, 左威健, 聂瑶瑶

太原科技大学计算机科学与技术学院, 太原 030024

收稿日期:2020-06-02 修回日期:2020-08-21 出版日期:2021-02-10 发布日期:2020-09-15
通讯作者: 胡立华
作者简介:胡立华(1982-),女,山西忻州人,副教授,博士,CCF会员,主要研究方向:计算机视觉、人工智能、模式识别;左威健(1995-),男,山西大同人,硕士研究生,主要研究方向:特征匹配、三维重建;聂瑶瑶(1995-),女,山西运城人,硕士研究生,主要研究方向:计算机视觉、三维重建。
基金资助:
国家自然科学基金资助项目（61873264）。

Feature matching method based on weighted similarity measurement

HU Lihua, ZUO Weijian, NIE Yaoyao

College of Computer Science and Technology, Taiyuan University of Science and Technology, Taiyuan Shanxi 030024, China

Received:2020-06-02 Revised:2020-08-21 Online:2021-02-10 Published:2020-09-15
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61873264).

摘要/Abstract

摘要： 针对图像特征匹配过程中采集图像易受噪声、光照、尺度等因素影响使产生的匹配结果鲁棒性差、误匹配率高等问题，提出一种基于加权相似性度量（WSM）的特征匹配方法。该方法首先采用基于网格多密度聚类的特征匹配（FM_GMC）算法对原始图像进行特征聚类块划分；其次在每一特征聚类块中，采用Canny提取边缘特征点并使用尺度不变特征变换（SIFT）进行描述；然后采用加权的方式对特征聚类块之间的空间上下文信息间的Hausdorff距离、图像特征点外观描述子间的欧氏距离以及图像特征点的局部几何灰度信息的归一化互相关度量（NCC）进行相似性度量；最后依据最近邻距离比值（NNDR）对相似性度量结果进一步优化，从而确定特征匹配结果。以古建筑图像为数据集的实验结果表明WSM方法的平均匹配精确率达到92%，在匹配数量和精确率上优于常用的特征匹配方法，验证了该方法的有效性和鲁棒性。

关键词: 特征匹配, 相似性度量, 尺度不变特征变换, Canny算法, 古建筑图像

Abstract: In order to solve the problems of poor robustness and high mismatch rate caused by the noise, illumination and scale in the image feature matching process, a feature matching method based on Weighted Similarity Measurement (WSM) was proposed. At first, FM_GMC (Feature Matching based on Grid and Multi-Density Clustering) algorithm was adopted to divide image into several feature clustering blocks. Secondly, in each feature clustering block, the edge feature points were attracted by Canny and descripted by Scale-Invariant Feature Transform (SIFT). Thirdly, the similarity measurements were performed on the Hausdorff distance of spatial context information between feature clustering blocks, the Euclidean distance between appearance descriptors of image feature points and Normalized Cross Correlation (NCC) by using weighting methods. Finally, the similarity measurement results were further optimized according to Nearest Neighbor Distance Ratio (NNDR), so as to determine the feature matching result. With the ancient architecture images as the dataset, the experimental results show that the WSM method has an average matching precision of 92%, and is superior to commonly matching algorithms on matching number and matching precision. Therefore, the effectiveness and robustness of WSM method are verified.

Key words: feature matching, similarity measurement, Scale Invariant Feature Transform (SIFT), Canny algorithm, ancient architecture image

中图分类号:

TP391.41

胡立华, 左威健, 聂瑶瑶. 基于加权相似性度量的特征匹配方法[J]. 计算机应用, 2021, 41(2): 511-516.

HU Lihua, ZUO Weijian, NIE Yaoyao. Feature matching method based on weighted similarity measurement[J]. Journal of Computer Applications, 2021, 41(2): 511-516.

参考文献

[1] FU J S,LIAN F L. Monocular vision-based collision-free navigation for flying robots[C]//Proceedings of the 8th International Congress on Advanced Applied Informatics. Piscataway:IEEE,2019:1091-1092.
[2] MUR-ARTAL R,MONTIEL J M M,TARDÓS J D. ORB-SLAM:a versatile and accurate monocular SLAM system[J]. IEEE Transactions on Robotics,2015,31(5):1147-1163.
[3] ZHANG Y,YE M,MANOCHA D,et al. 3D reconstruction in the presence of glass and mirrors by acoustic and visual fusion[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018,40(8):1785-1798.
[4] YANG B,ROSA S,MARKHAM A,et al. Dense 3D object reconstruction from a single depth view[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2019,41(12):2820-2834.
[5] WANG H,FENG L,ZHANG J,et al. Semantic discriminative metric learning for image similarity measurement[J]. IEEE Transactions on Multimedia,2016,18(8):1579-1589.
[6] 周菊香. 图像检索中的特征表达和相似性度量方法研究[D]. 大连:大连理工大学,2019:22-45. (ZHOU J X. Feature representation and similarity metric in image retrieval[D]. Dalian:Dalian University of Technology,2019:22-45.)
[7] ZHAO C,SHI W,DENG Y. A new Hausdorff distance for image matching[J]. Pattern Recognition Letters,2005,26(5):581-586.
[8] WANG X,FU W P,WANG X. Optimized SIFT image matching algorithm[C]//Proceedings of the 2008 International Conference on Automation and Logistics. Piscataway:IEEE,2008:843-847.
[9] KANG J,QI N,HOU J. A hybrid method combining Hausdorff distance,genetic algorithm and simulated annealing algorithm for image matching[C]//Proceedings of the 2nd International Conference on Computer Modeling and Simulation. Piscataway:IEEE,2010:435-439.
[10] 刘坡, 张宇, 龚建华. 中误差和邻近关系的多尺度面实体匹配算法研究[J]. 测绘学报,2014,43(4):419-425.(LIU P, ZHANG Y,GONG J H. Root mean square error and neighbouring relation matching approach for multi-scale areal feature[J]. Acta Geodaetica Et Cartographica Sinica,2014,43(4):419-425.)
[11] KUMAR K S, ELAVARASAN S. Shape matching and hand written digit identification using Minkowski distance[J]. Biometrics and Bioinformatics,2017,9(6):116-118.
[12] 张明英. 图像特征相似性度量方法研究及应用[D]. 成都:电子科技大学, 2019:15-45. (ZHANG M Y. Similarity measurement of image feature and its application[D]. Chengdu:University of Electronic Science and Technology of China,2019:15-45.)
[13] WEI S D,LAI S H. Fast template matching based on normalized cross correlation with adaptive multilevel winner update[J]. IEEE Transactions on Image Processing,2008,17(11):2227-2235.
[14] MAHMOOD A, KHAN S. Correlation-coefficient-based fast template matching through partial elimination[J]. IEEE Transactions on Image Processing,2012,21(4):2099-2108.
[15] KOLAR R,HARABIS V,ODSTRCILIC J. Hybrid retinal image registration using phase correlation[J]. The Imaging Science Journal,2013,61(4):369-384.
[16] WOO J,STONE M,PRINCE J L. Multimodal registration via mutual information incorporating geometric and spatial context[J]. IEEE Transactions on Image Processing,2015,24(2):757-769.
[17] LIN L,WANG G,ZUO W,et al. Cross-domain visual matching via generalized similarity measure and feature learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2017, 39(6):1089-1102.
[18] RUBNER Y,TOMASI C,GUIBAS L J. The earth mover's distance as a metric for image retrieval[J]. International Journal of Computer Vision,2000,40(2):99-121.
[19] YOON K J,KWEON I S. Stereo matching with the distinctive similarity measure[C]//Proceedings of the IEEE 11th International Conference on Computer Vision. Piscataway:IEEE, 2007:1-7.
[20] WONG A,CLAUSI D A,FIEGUTH P. CPOL:complex phase order likelihood as a similarity measure for MR-CT registration[J]. Medical Image Analysis,2010,14(1):50-57.
[21] ZHOU Z, QIN B. Robust non-rigid image registration using incomplete information, adaptive normalized convolution and similarity measure combination[C]//Proceedings of the 2007 IEEE Symposium on Nuclear Science. Piscataway:IEEE,2007:3197-3201.
[22] GUO J,FAN Y,TAN H,et al. Image matching using structural similarity and geometric constraint approaches on remote sensing images[J]. Journal of Applied Remote Sensing,2016,10(4):No. 045007.
[23] LIN Z,LYU M R,KING I. MatchSim:a novel similarity measure based on maximum neighborhood matching[J]. Knowledge and Information Systems,2012,32(1):141-166.
[24] 聂瑶瑶, 胡立华, 张继福, 等. 基于网格多密度的古建筑图像特征匹配方法[J]. 计算机辅助设计与图形学学报,2020,32(3):437-444. (NIE Y Y,HU L H,ZHANG J F,et al. Feature matching based on the grid and multi-density for ancient architectural images[J]. Journal of Computer-Aided Design and Computer Graphics,2020,32(3):437-444.)
[25] MIKOLAJCZYK K, SCHMID C. A performance evaluation of local descriptors[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(10):1615-1630.

基于加权相似性度量的特征匹配方法

Feature matching method based on weighted similarity measurement

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	杨蒙蒙, 张爱华. 基于灰度共生矩阵和同步正交匹配追踪的分形图像压缩[J]. 计算机应用, 2021, 41(5): 1445-1449.
[2]	欧莉莉, 邵峰晶, 孙仁诚, 隋毅. 基于半监督方法的脑梗死图像识别[J]. 计算机应用, 2021, 41(4): 1221-1226.
[3]	马伟苹, 李文新, 孙晋川, 曹鹏霞. 基于粗精立体匹配的双目视觉目标定位方法[J]. 计算机应用, 2020, 40(1): 227-232.
[4]	章永来, 周耀鉴. 聚类算法综述[J]. 计算机应用, 2019, 39(7): 1869-1882.
[5]	薄单, 李宗春, 王晓南, 乔涵文. 适用于倾斜影像的加速KAZE-SIFT特征提取算法[J]. 计算机应用, 2019, 39(7): 2093-2097.
[6]	李南云, 王旭光, 吴华强, 何青林. 基于灰度塔评分的匹配模型构建在无人机网络视频拼接中的应用[J]. 计算机应用, 2019, 39(5): 1480-1484.
[7]	姜逸凡, 叶青. 基于孪生神经网络的时间序列相似性度量[J]. 计算机应用, 2019, 39(4): 1041-1045.
[8]	马兰, 王京杰, 陈焕. 基于内容挖掘的广域信息管理系统业务数据安全[J]. 计算机应用, 2019, 39(2): 488-493.
[9]	杨春德, 成燕菲. 基于拼接缝自适应消除和全景图矫直的快速图像拼接算法[J]. 计算机应用, 2019, 39(10): 3053-3059.
[10]	李加亮, 蒋品群. 结合变形函数和幂函数权重的图像拼接[J]. 计算机应用, 2019, 39(10): 3060-3064.
[11]	陆明炽, 王守华, 李云柯, 纪元法, 孙希延, 邓桂辉. 基于特征匹配和距离加权的蓝牙定位算法[J]. 计算机应用, 2018, 38(8): 2359-2364.
[12]	常炳国, 臧虹颖. 基于分段降维和路径修正DTW的时序特征分类器设计[J]. 计算机应用, 2018, 38(7): 1910-1915.
[13]	鲍舒婷, 孙丽萍, 郑孝遥, 郭良敏. 基于共享近邻相似度的密度峰聚类算法[J]. 计算机应用, 2018, 38(6): 1601-1607.
[14]	陈虹, 肖越, 肖成龙, 宋好. 基于SIFT算子融合最大相异系数的自适应图像匹配算法[J]. 计算机应用, 2018, 38(5): 1410-1414.
[15]	叶峰, 洪峥, 赖乙宗, 赵雨亭, 谢先治. 基于匹配质量提纯的改进D-Nets算法[J]. 计算机应用, 2018, 38(4): 1121-1126.