图论算法的无人机影像匹配特征点粗差剔除

doi:10.13474/j.cnki.11-2246.2020.0103

测绘通报 ›› 2020, Vol. 0 ›› Issue (4): 6-10.doi: 10.13474/j.cnki.11-2246.2020.0103

图论算法的无人机影像匹配特征点粗差剔除

喜文飞¹, 史正涛¹, 李国柱²

1. 云南师范大学旅游与地理科学学院, 云南昆明 650500;
2. 云南海钜地理信息技术有限公司, 云南昆明 650000

收稿日期:2019-07-09 修回日期:2020-02-29 出版日期:2020-04-25 发布日期:2020-05-08
作者简介:喜文飞(1984-),男,博士生,主要研究无人机图像处理、三维模型构建、变形监测。E-mail:xiwenfei911@163.com
基金资助:
云南省科技厅面上项目（2017FB078）；云南师范大学教改项目（YNJG201845）

UAV image matching feature point coarse error elimination based on image theory algorithm

XI Wenfei¹, SHI Zhengtao¹, LI Guozhu²

1. College of Tourism and Geographic Sciences, Yunnan Normal University, Kunming 650500, China;
2. Yunnan Haiju Geographic Information Technology Co., Ltd., Kunming 650000, China

Received:2019-07-09 Revised:2020-02-29 Online:2020-04-25 Published:2020-05-08

摘要/Abstract

摘要： 无人机影像匹配过程中，粗差是不可避免的，因此，获取稳健性较高的特征点进行无人机影像匹配至关重要。传统的方法是采用经典的RANSAC算法进行粗差剔除，该算法受抽样次数、误差阈值的影响，还会残存部分误匹配的特征点。利用图论原理，对SIFT算法提取的特征点进行预处理，通过构建特征点的能量函数剔除能量较低的特征点，可以提高匹配特征点的稳健性，减少特征点的粗差。本文提出了一种新的方法，将图论算法与经典的RANSAC算法相结合进行粗差剔除。该方法命名为GSIFT-RANSAC算法，利用该算法可以提高特征点的稳健性，获取高精度的单应矩阵。采用两组无人机影像进行验证，本文提出的算法与单独利用图论剔除特征点的算法相比，粗差剔除率分别提高了5.31%和14.29%，说明该方法效果较好。

关键词: 无人机影像匹配, 特征点, RANSAC算法, 图论, GSIFT-RANSAC算法

Abstract: In the process of UAV image matching, rough error is inevitable. Therefore, it is very important to obtain feature points with high robustness for UAV image matching. The traditional method is using classical RANSAC algorithm for coarse error elimination, which is affected by the sampling times, error threshold, and remaining partial mismatched points. By using graph theory, the feature points extracted by SIFT algorithm are preprocessed, that is the feature points with lower energy are removed by constructing the energy function of feature points, which can improve the robustness of matching feature points and reduce the feature points coarse error. The paper proposes a new method, which combines the graph theory algorithm with the classical RANSAC algorithm to eliminate the rough error. The method is named GSIFT-RANSAC algorithm, which can improve the robustness of feature points and obtain the homography matrix with high accuracy. Using different data for verification, gross error elimination rate of the algorithm proposed in this paper is 5.31% and 14.29% higher than the algorithm using graph theory to remove feature points alone, which indicates that the effect of proposed method is better.

Key words: UAV image matching, feature points, RANSAC algorithm, graph theory, GSIFT-RANSAC algorithm

中图分类号:

P237

喜文飞, 史正涛, 李国柱. 图论算法的无人机影像匹配特征点粗差剔除[J]. 测绘通报, 2020, 0(4): 6-10.

XI Wenfei, SHI Zhengtao, LI Guozhu. UAV image matching feature point coarse error elimination based on image theory algorithm[J]. Bulletin of Surveying and Mapping, 2020, 0(4): 6-10.

参考文献

[1] XING Y J, MENG J J,SUN J,et al.An improved region growth algorithm for dense matching[J]. Journal of Achievements in Materials and Manufacturing Engineering,2006, 18(1-2):323-326.
[2] 邓宝松,高宇,魏迎梅,等.一种评价仿射不变性特征定位误差的新方法[J].中国图象图形学报,2008,13(2):291-297.
[3] LI J, ALLINSON N M. A comprehensive review of current local features for computer vision[J].Neurocomputing,2008, 71(10-12):1771-1787.
[4] MIKOLAJCZYK K,TUYELAARS T,SCHMID C,et al. A comparison of affine region detectors[J]. International Journal of Computer Vision,2005,65(2):43-72.
[5] MEGYESI Z. Dense matching methods for 3D scene reconstruc-tion from wide baseline images[D]. France:Eotvos Lorand University,2009.
[6] 郭建锋,赵俊.粗差探测与识别统计检验量的比较分析[J].测绘学报,2012,5(1):14-18.
[7] 刘进,闫利.图像相关匹配算法的快速实现[J].武汉大学学报(信息科学版),2007,32(8):684-687.
[8] LI J Y, HU Q W, AI M Y,ZHONG R F. Robust feature matching via support-line voting and affine-invariantratios[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2017(132):61-76.
[9] STEWART C V. A new robust operator for computer vision:theoretical analysis[C]//Proceedings of Computer Vision and Pattern Recognition. Seattle:IEEE, 1994:167-173.
[10] TORR P H S,ZISSERMAN A. MLESAC:A new robust estimator with application to estimating image geometry[J]. Computer Vision & Image Understanding,2000,78(1):138-156.
[11] CHUM O,MATAS J,OBDRZALEK S. Enhancing RANSAC by generalized model optimization[EB/OL].2017-05-15[2019-06-15]. http://cmp.felk.cvut.cz/ftp/articles/chum/chum-accv04.pdf.
[12] SEDAGHAT A,EBADI H,MOKHTARZADE M. Image matching of satellite data based on quadrilateral control networks[J]. Photogrammetric Record,2012,27(140):423-442.
[13] 张永军,王博,黄旭,等.影像匹配粗差的局部矢量面元剔除方法[J].测绘学报,2014,43(7):717-723.
[14] 李智强.资源卫星影像匹配粗差筛除算法设计与实现[J].测绘科学,2017,42(6):202-206.
[15] 郭莹,牛继强,陈时雨,等.基于图论的影像匹配粗差探测算法[J].测绘科学,2018,43(11):85-91.
[16] 伏雪, 马燕,林涛.一种新的基于图论的图像匹配算法[J].计算机应用与软件,2016,33(12):156-159.
[17] LOWE D G. Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[18] FISHIER M A, BOLES R C. Random sample concensus:a paradigm for model fitting with applications to image analysis and automated cartography[J].Communications of the ACM,1981,24(6):381-395.
[19] 宋卫艳.RANSAC算法及其在遥感图像处理中的应用[D].北京:华北电力大学,2011.
[20] 李明哲,金俊,石端银.图论及其算法[M]. 北京:机械工业出版社,2016.

图论算法的无人机影像匹配特征点粗差剔除

UAV image matching feature point coarse error elimination based on image theory algorithm

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

本文评价

[1]	王贝, 叶险峰, 王军宁. 利用点云切片提取技术分析桥梁形变[J]. 测绘通报, 2022, 0(4): 167-171.
[2]	李国竣, 徐延海, 段杰文, 韩石磊. 利用局部自适应阈值方法提取ORB-SLAM特征点[J]. 测绘通报, 2021, 0(9): 32-36,48.
[3]	韩飞落. 一种用于BIM模型坐标系转换至工程坐标系的方法[J]. 测绘通报, 2021, 0(6): 163-165.
[4]	王挺, 陈超然, 彭梦圆, 黄尔双, 黄霞. POS辅助和核线约束的倾斜影像匹配方法[J]. 测绘通报, 2021, 0(5): 91-95.
[5]	喜文飞, 赵子龙, 王绍君, 沈志坚, 李帅. 一种改进的无人机影像拼接粗差剔除算法[J]. 测绘通报, 2021, 0(2): 36-39.
[6]	张少平, 翟永聪, 王留召, 毛明楷. 车载激光测量技术在质检应用中的可行性[J]. 测绘通报, 2021, 0(12): 110-114.
[7]	何志伟, 唐伯惠, 王涛, 王晓红, 于伯华, 李闯, 邓仕雄. 剔除无人机影像BRISK特征误匹配点对算法[J]. 测绘通报, 2021, 0(10): 78-82.
[8]	周杰, 解琨, 付超, 施昆. 结合SIFT特征点和泊松融合的无人机遥感影像拼接技术[J]. 测绘通报, 2021, 0(1): 94-98.
[9]	薛慧艳, 王智. 基于线定向设站法建立独立坐标系的探讨及其应用[J]. 测绘通报, 2020, 0(9): 155-158.
[10]	李志远, 王健, 于德亮, 孙文潇. 基于点云的地铁隧道断面精确拟合与形变分析[J]. 测绘通报, 2020, 0(8): 157-159.
[11]	聂运菊, 熊佳诚, 程朋根, 罗跃. 结合PS特征点的SBAS地表形变监测[J]. 测绘通报, 2020, 0(4): 91-95.
[12]	张冬梅, 卢小平, 苗沛基, 周雨石, 马靓婷. 一种改进的APAP影像匹配算法[J]. 测绘通报, 2020, 0(3): 123-128.
[13]	张进, 赵相伟, 栾吉山, 冯康, 艾波. 改进FAST和对立颜色特征的向量场一致性匹配[J]. 测绘通报, 2020, 0(11): 50-54.
[14]	黄炅怡, 黄培之, 李坚强. 利用流水矢量提取山脊线和山谷线特征点的方法[J]. 测绘通报, 2020, 0(10): 12-15.
[15]	张鸿刚, 李成名, 于美娜, 武鹏达, 郭曼. 多特征约束下的复杂交叉路口主辅路识别[J]. 测绘通报, 2020, 0(10): 21-25,42.