面向无人机倾斜影像的高效SfM重建方案
|
摘要
针对无人机影像分辨率高、数据量大导致稀疏重建效率低的问题,提出了一种减少影像匹配对数量、提高外点剔除效率的算法。首先利用无人机飞控数据、相机安装角计算影像的粗略POS(positioning and orientation system)信息;然后基于拓扑连接分析设计了最大生成树算法(maximum spanning tree expansion,MST-Expansion),最大程度地降低影像配对数;考虑到初始匹配的高外点率,设计了分层运动一致性约束(hierarchical motion consistency constraint, HMCC)算法,提高几何验证算法的效率。4组不同倾斜设备采集的无人机影像的实验结果说明,该方案能够在保证重建精度的前提下,实现无人机倾斜影像高效和稳健的稀疏重建。
For the low sparse reconstruction efficiency caused by high resolutions and large volumes of UAV(unmanned aerial vehicle) images, this paper proposes an algorithm for decreasing the number of image pairs and improving the efficiency of outlier removal. Firstly, rough POS(positioning and orientation system) is calculated for each image with the use of GNSS/IMU(Global Navigation Satellite System/inertial measurement unit) data and camera installation angles. Secondly, to reduce image combination complexity, topological connection analysis is used for image pairs selection. Considering high outlier ratios of initial matches, the hierarchical motion consistency constraint(HMCC) is designed to achieve the high efficiency of geometrical verification strategies. The proposed solutions are verified by using four datasets captured with different oblique systems. Results demonstrate that without accuracy sacrifices, the proposed solutions can achieve efficient and reliable reconstruction.
For the low sparse reconstruction efficiency caused by high resolutions and large volumes of UAV(unmanned aerial vehicle) images, this paper proposes an algorithm for decreasing the number of image pairs and improving the efficiency of outlier removal. Firstly, rough POS(positioning and orientation system) is calculated for each image with the use of GNSS/IMU(Global Navigation Satellite System/inertial measurement unit) data and camera installation angles. Secondly, to reduce image combination complexity, topological connection analysis is used for image pairs selection. Considering high outlier ratios of initial matches, the hierarchical motion consistency constraint(HMCC) is designed to achieve the high efficiency of geometrical verification strategies. The proposed solutions are verified by using four datasets captured with different oblique systems. Results demonstrate that without accuracy sacrifices, the proposed solutions can achieve efficient and reliable reconstruction.
引文
[1]Li Deren,Li Ming.Research Advance and Application Prospect of Unmanned Aerial Vehicle Remote Sensing System[J].Geomatics and Information Science of Wuhan University,2014,39(5):505-513(李德仁,李明.无人机遥感系统的研究进展与应用前景[J].武汉大学学报·信息科学版,2014,39(5):505-513)
[2]Li Pengfei,Sun Kaimin,Li Deren,et al.A Load Balancing Strategy for Urgent Parallel Processing of UAV Imagery[J].Geomatics and Information Science of Wuhan University,2018,43(2):268-274(李鹏飞,孙开敏,李德仁,等.无人机影像应急并行处理负载均衡方法[J].武汉大学学报·信息科学版,2018,43(2):268-274)
[3]Snavely N,Seitz S M,Szeliski R.Photo Tourism:Exploring Photo Collections in 3D[J].ACM Transactions on Graphics,2006,25(3):835-846
[4]Hu H,Zhu Q,Du Z,et al.Reliable Spatial Relationship Constrained Feature Point Matching of Oblique Aerial Images[J].Photogrammetric Engineering and Remote Sensing,2015,81(1):49-58
[5]Lu L,Zang Y,Tao P.Geometrical Consistency Voting Strategy for Outlier Detection in Image Matching[J].Photogrammetric Engineering and Remote Sensing,2016,82(7):559-570
[6]Rupnik E,Nex F,Remondino F.Automatic Orientation of Large Blocks of Oblique Images[C].International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,Hannover,Germany,2013
[7]Xu Z,Wu L,Gerke M,et al.Skeletal Camera Network Embedded Structure-from-Motion for 3D Scene Reconstruction from UAV Images[J].ISPRS Journal of Photogrammetry and Remote Sensing,2016,121:113-127
[8]Chum O,Matas J.Optimal Randomized RANSAC[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(8):1 472-1 482
[9]Li X,Larson M,Hanjalic A.Pairwise Geometric Matching for Large-Scale Object Retrieval[C].IEEE Conference on Computer Vision and Pattern Recognition,Boston,USA,2015
[10]Sch?nberger J L,Price T,Sattler T,et al.A Voteand-Verify Strategy for Fast Spatial Verification in Image Retrieval[C].Asian Conference on Computer Vision,Taipei,China,2016
[11]Sattler T,Leibe B,Kobbelt L.SCRAMSAC:Improving RANSAC’s Efficiency with a Spatial Consistency Filter[C].International Conference on Computer Vision,Kyoto,Japan,2010
[12]Jiang S,Jiang W.On-Board GNSS/IMU Assisted Feature Extraction and Matching for Oblique UAVImages[J].Remote Sensing,2017,9(8):813-833
[13]Graham R L,Heill P.On the History of the Minimum Spanning Tree Problem[J].Annals of the History of Computing,1985,7(1):43-57
[14]Kruskal J B.On the Shortest Spanning Subtree of a Graph and the Traveling Salesman Problem[J].Proceedings of the American Mathematical Society,1956,7(1):48-50
[15]Wu C.SiftGPU:A GPU Implementation of Scale Invariant Feature Transform(SIFT)[OL].https://github.com/pitzer/SiftGPU,2017
[16]Sun Y,Zhao L,Huang S,et al.L2-SIFT:SIFTFeature Extraction and Matching for Large Images in Large-Scale Aerial Photogrammetry[J].ISPRSJournal of Photogrammetry and Remote Sensing,2014,91:1-16
[17]Sameer A,Keir M.Ceres Solver[OL].http://ceres-solver.org,2018
[18]Chum O,Matas J,Kittler J.Locally Optimized RANSAC[C].Pattern Recognition,the 25th DAGMSymposium,Magdeburg,Germany,2003
[2]Li Pengfei,Sun Kaimin,Li Deren,et al.A Load Balancing Strategy for Urgent Parallel Processing of UAV Imagery[J].Geomatics and Information Science of Wuhan University,2018,43(2):268-274(李鹏飞,孙开敏,李德仁,等.无人机影像应急并行处理负载均衡方法[J].武汉大学学报·信息科学版,2018,43(2):268-274)
[3]Snavely N,Seitz S M,Szeliski R.Photo Tourism:Exploring Photo Collections in 3D[J].ACM Transactions on Graphics,2006,25(3):835-846
[4]Hu H,Zhu Q,Du Z,et al.Reliable Spatial Relationship Constrained Feature Point Matching of Oblique Aerial Images[J].Photogrammetric Engineering and Remote Sensing,2015,81(1):49-58
[5]Lu L,Zang Y,Tao P.Geometrical Consistency Voting Strategy for Outlier Detection in Image Matching[J].Photogrammetric Engineering and Remote Sensing,2016,82(7):559-570
[6]Rupnik E,Nex F,Remondino F.Automatic Orientation of Large Blocks of Oblique Images[C].International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,Hannover,Germany,2013
[7]Xu Z,Wu L,Gerke M,et al.Skeletal Camera Network Embedded Structure-from-Motion for 3D Scene Reconstruction from UAV Images[J].ISPRS Journal of Photogrammetry and Remote Sensing,2016,121:113-127
[8]Chum O,Matas J.Optimal Randomized RANSAC[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(8):1 472-1 482
[9]Li X,Larson M,Hanjalic A.Pairwise Geometric Matching for Large-Scale Object Retrieval[C].IEEE Conference on Computer Vision and Pattern Recognition,Boston,USA,2015
[10]Sch?nberger J L,Price T,Sattler T,et al.A Voteand-Verify Strategy for Fast Spatial Verification in Image Retrieval[C].Asian Conference on Computer Vision,Taipei,China,2016
[11]Sattler T,Leibe B,Kobbelt L.SCRAMSAC:Improving RANSAC’s Efficiency with a Spatial Consistency Filter[C].International Conference on Computer Vision,Kyoto,Japan,2010
[12]Jiang S,Jiang W.On-Board GNSS/IMU Assisted Feature Extraction and Matching for Oblique UAVImages[J].Remote Sensing,2017,9(8):813-833
[13]Graham R L,Heill P.On the History of the Minimum Spanning Tree Problem[J].Annals of the History of Computing,1985,7(1):43-57
[14]Kruskal J B.On the Shortest Spanning Subtree of a Graph and the Traveling Salesman Problem[J].Proceedings of the American Mathematical Society,1956,7(1):48-50
[15]Wu C.SiftGPU:A GPU Implementation of Scale Invariant Feature Transform(SIFT)[OL].https://github.com/pitzer/SiftGPU,2017
[16]Sun Y,Zhao L,Huang S,et al.L2-SIFT:SIFTFeature Extraction and Matching for Large Images in Large-Scale Aerial Photogrammetry[J].ISPRSJournal of Photogrammetry and Remote Sensing,2014,91:1-16
[17]Sameer A,Keir M.Ceres Solver[OL].http://ceres-solver.org,2018
[18]Chum O,Matas J,Kittler J.Locally Optimized RANSAC[C].Pattern Recognition,the 25th DAGMSymposium,Magdeburg,Germany,2003