|Table of Contents|

Rapid Pose Estimation and Emergency Mapping for UAV Images Based on Common-view Optimization in GNSS-denied Environments(PDF)

《地球科学与环境学报》[ISSN:1672-6561/CN:61-1423/P]

Issue:
2025年第05期
Page:
987-998
Research Field:
大地测量、遥感与地学大数据
Publishing date:

Info

Title:
Rapid Pose Estimation and Emergency Mapping for UAV Images Based on Common-view Optimization in GNSS-denied Environments
Author(s):
GU Di-zhen1;  YANG Yun1; 2*;  ZHAO Bo3;  LI Zu-feng4;  HAO Guo-pu1;  CHEN Shi-chang1;  YANG Cheng-sheng1;  TANG Yi-liang1
(1. School of Geological Engineering and Geomatics, Chang'an University, Xi'an 710054, Shaanxi, China; 2. State Key Laboratory of Geo-information Engineering, Xi'an 710054, Shaanxi, China; 3. Surveying and Exploration Institute of Geology and Mineral Exploration and Development of Gansu Provincial Bureau, Lanzhou 730060, Gansu, China; 4. Northwest Engineering Corporation Limited, PowerChina, Xi'an 710065, Shaanxi, China)
Keywords:
UAV image sequence;  emergency scene;  image retrieval;  common-view graph;  feature matching;  pose estimation;  structure from motion
PACS:
P231
DOI:
10.19814/j.jese.2025.01042
Abstract:
In order to improve the efficiency of UAV image pose estimation in GNSS-denied environments, a fast mapping algorithm based on common-view graph optimization is proposed, named as NSG-VLAD. First, scale invariant feature transform(SIFT)is employed to extract features from each image, and then the vector of locally aggregated descriptors(VLAD)algorithm is used to aggregate the feature descriptor into a global feature vector; subsequently, the graph-based approximate nearest neighbor search(ANNS)algorithm is utilized for retrieving similar images; finally, a common-view graph is constructed for each pair of similar images, followed by iterative matching, thereby enhancing the efficiency of image feature matching, pose estimation, and mapping in emergency scence. By utilizing NPU_FACTORY, NPU_PARK and three self-created datasets, NSG-VLAD algorithm is compared with the representative Colmap open-source software, as well as Metaphase and Pix4Dmapper commercial softwares. The results show that NSG-VLAD algorithm is 3 times higher than Metashape commercial software in image reconstruction, and 10 times faster than Colmap open-source software; re-projection error of NSG-VLAD algorithm surpasses that of Colmap open-source software and Metashape commercial software. The 3D point cloud mapping speed is at least 2 times faster than the similar methods in emergency surveying tasks, validating that NSG-VLAD algorithm has a good application prospect on emergency mapping in GNSS-denied environments such as earthquake disaster areas.

References:

[1] 陈 武,姜 三,李清泉,等.无人机影像增量式运动恢复结构研究进展[J].武汉大学学报(信息科学版),2022,47(10):1662-1674.
CHEN Wu,JIANG San,LI Qing-quan,et al.Recent Research of Incremental Structure from Motion for Unmanned Aerial Vehicle Images[J].Geomatics and Information Science of Wuhan University,2022,47(10):1662-1674.
[2] JIANG S,JIANG W S,WANG L Z.Unmanned Aerial Vehicle-based Photogrammetric 3D Mapping:A Survey of Techniques,Applications,and Challenges[J].IEEE Geoscience and Remote Sensing Magazine,2022,10(2):135-171.
[3] 李龙威.无人机航摄影像快速正射拼接及三维重建系统设计与实现[D].西安:长安大学,2023.
LI Long-wei.Design and Implementation of Fast Orthophoto Mosaic and 3D Reconstruction System for UAV Aerial Images[D].Xi'an:Chang'an University,2023.
[4] 高 翔,李梦晗,申抒含.大规模场景运动恢复结构研究综述[J].计算机辅助设计与图形学学报,2024,36(7):969-994.
GAO Xiang,LI Meng-han,SHEN Shu-han.Large-scale Structure from Motion:A Survey[J].Journal of Computer-aided Design & Computer Graphics,2024,36(7):969-994.
[5] SNAVELY N,SEITZ S M,SZELISKI R.Photo Tou-rism:Exploring Photo Collections in 3D[J].ACM Transactions on Graphics,2006,25(3):835-846.
[6] SCHÖNBERGER J L,FRAHM J M.Structure-from-motion Revisited[C]∥IEEE.2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Las Vegas:IEEE,2016:4104-4113.
[7] 任超锋,蒲禹池,张福强.顾及地理空间信息的无人机影像匹配像对提取方法[J].自然资源遥感,2022,34(1):85-92.
REN Chao-feng,PU Yu-chi,ZHANG Fu-qiang.A Method for Extracting Match Pairs of UAV Images Considering Geospatial Information[J].Remote Sensing for Natural Resources,2022,34(1):85-92.
[8] 姜 三,许志海,张 峰,等.面向无人机倾斜影像的高效SfM重建方案[J].武汉大学学报(信息科学版),2019,44(8):1153-1161.
JIANG San,XU Zhi-hai,ZHANG Feng,et al.Solution for Efficient SfM Reconstruction of Oblique UAV Images[J].Geomatics and Information Science of Wuhan University,2019,44(8):1153-1161.
[9] 陈怀圆,党建武,岳 彪,等.基于并行化处理的无人机影像三维重建算法[J].激光与光电子学进展,2024,61(8):109-117.
CHEN Huai-yuan,DANG Jian-wu,YUE Biao,et al.Three Dimensional Reconstruction Algorithm of Unmanned Aerial Vehicle Images Based on Parallel Processing[J].Laser & Optoelectronics Progress,2024,61(8):109-117.
[10] 姜 三,江万寿,郭丙轩.词汇树索引约束的无人机影像快速特征匹配算法[J].武汉大学学报(信息科学版),2024,49(9):1597-1609.
JIANG San,JIANG Wan-shou,GUO Bing-xuan.Fast Feature Matching of UAV Images via Indexing Constraints of Vocabulary Trees[J].Geomatics and Information Science of Wuhan University,2024,49(9):1597-1609.
[11] 刘思康,郭丙轩,姜 三,等.图索引结构词袋模型的无人机影像匹配对检索[J].测绘通报,2023(4):93-98.
LIU Si-kang,GUO Bing-xuan,JIANG San,et al.Mat-ching Pair Retrieval Method of UAV Images Based on the Graph Structure Bag of Words Model[J].Bulletin of Surveying and Mapping,2023(4):93-98.
[12] SCHÖNBERGER J L,PRICE T,SATTLER T,et al.A Vote-and-verify Strategy for Fast Spatial Verification in Image Retrieval[C]∥LAI S H,LEPETIT V,NISHINO K,et al.Computer Vision—ACCV 2016:13th Asian Conference on Computer Vision.Cham:Springer,2016:321-337.
[13] 李一德.基于无序图像集的非结构化场景恢复关键技术研究[D].西安:西安工业大学,2024.
LI Yi-de.Research on Key Technologies for Unstructured Scene Restoration Based on Unordered Image Sets[D].Xi'an:Xi'an Technological University,2024.
[14] LIU J H,MA Y C,JIANG S,et al.Matchable Image Retrieval for Large-scale UAV Images:An Evaluation of SfM-based Reconstruction[J].International Journal of Remote Sensing,2024,45(3):692-718.
[15] BRUCH S.Graph Algorithms[M]∥BRUCH S.Foundations of Vector Retrieval.Amsterdam:Springer,2024:73-103.
[16] WANG M Z,XU X L,YUE Q,et al.A Comprehensive Survey and Experimental Comparison of Graph-based Approximate Nearest Neighbor Search[J].Proceedings of the VLDB Endowment,2021,14(11):1964-1978.
[17] HALDER R K,UDDIN M N,UDDIN M A,et al.Enhancing K-nearest Neighbor Algorithm:A Comprehensive Review and Performance Analysis of Modifications[J].Journal of Big Data,2024,11(1):113.
[18] YE Z C,ZHANG G F,BAO H J.Efficient Covisibility-based Image Matching for Large-scale SfM[C]∥IEEE.2020 IEEE International Conference on Robotics and Automation(ICRA).Paris:IEEE,2020:8616-8622.
[19] 姜 三,马一尘,李清泉,等.无序无人机影像的并行化SfM三维重建方法[J].测绘学报,2024,53(5):946-958.
JIANG San,MA Yi-chen,LI Qing-quan,et al.Parallel SfM-based 3D Reconstruction for Unordered UAV Images[J].Acta Geodaetica et Cartographica Sinica,2024,53(5):946-958.
[20] WANG Z,SHI D X,QIU C P,et al.Sequence Match-ing for Image-based UAV-to-satellite Geolocalization[J].IEEE Transactions on Geoscience and Remote Sensing,2024,62:5607815.
[21] BU S H,ZHAO Y,WAN G,et al.Map2DFusion:Real-time Incremental UAV Image Mosaicing Based on Monocular SLAM[C]∥IEEE.2016 IEEE/RSJ International Conference on Intelligent Robots and Systems(IROS).Daejeon:IEEE,2016:4564-4571.
[22] JARAHIZADEH S,SALEHI B.A Comparative Analysis of UAV Photogrammetric Software Performance for Forest 3D Modeling:A Case Study Using AgiSoft Photoscan,PIX4DMapper,and DJI Terra[J].Sensors,2024,24(1):286.
[23] YE Z C,BAO C,ZHOU X,et al.EC-SfM:Efficient Covisibility-based Structure-from-motion for Both Sequential and Unordered Images[J].IEEE Transactions on Circuits and Systems for Video Technology,2024,34(1):110-123.

Memo

Memo:
-
Last Update: 2025-10-01