A Real-time Unmanned Aerial Vehicle Localization Method Based on Multi-view Stereo Vision

doi:10.13474/j.cnki.11-2246.2018.0106

Abstract

Abstract:

In this paper,the author proposes a real-time unmanned aerial vehicle auto localization method based on multi-view stereo vision.Firstly,cameras are divided into several groups,and each group is connected with a computer to construct a local area network.Then,interior and exterior elements as well as the lens distortion parameters are measured.Next,images are captured simultaneously and the unmanned aerial vehicle is detected and located automatically by background difference and forward intersection.In order to verify this method,an unmanned aerial vehicle localization system is designed and implemented.The experimental results show that the method is feasible and effective and the expected accuracy is reached.

Key words: unmanned aerial vehicle auto localization, local area network construction, multi-view stereo vision, background difference

CLC Number:

P237

WANG Chen, ZHENG Shunyi, ZHU Fengbo, GUI Li. A Real-time Unmanned Aerial Vehicle Localization Method Based on Multi-view Stereo Vision[J]. 测绘通报, 2018, 0(4): 32-35,82.

References

[1] 解邦福.基于单目视觉的刚体位姿测量系统研究[D].哈尔滨:哈尔滨工业大学,2009.
[2] BENALLAL M,MEUNIER J.Pose Estimation of Polygonal Object in Monocular Vision Using Parametric Equations of Vertices[C]//IAPR Conference on Machine Vision Applications.[S.l.]:DBLP,2002:342-345.
[3] LIU C,ZHANG Y.3D Trajectory Reconstruction from Monocular Vision Based on Prior Spatial Knowledge[J].IEEE Sensors Journal,2016,16(3):817-822.
[4] JUNG S,JANG D,KIM M.An Approach to 3D Object Localization Based on Monocular Vision[J].Journal of Korea Multimedia Society,2008,11:1658-1667.
[5] XU S,LIU M.Feature Selection and Pose Estimation from Known Planar Objects Using Monocular Vision[C]//IEEE International Conference on Robotics and Biomimetics.Shenzhen:IEEE,2014:922-927.
[6] DOIGNON C,ABBA G,OSTERTAG E.Recognition and Localization of Solid Objects by a Monocular Vision System for Robotic Tasks[J].Intelligent Robots & Systems Advanced Robotic Systems & the Real World Ir,1994(3):2007-2014.
[7] HOERMANN S,BORGES P V K.Robot Localization Using 3D-models and an Off-board Monocular Camera[C]//IEEE International Conference on Computer Vision Workshops.Barcelona:IEEE,2011:1006-1013.
[8] LIU W,WANG Z,NING X,et al.An Object Recognition and Location Based on Binocular Stereo Vision[C]//International Forum on Computer Science-Technology and Applications.Chongqing:IEEE Computer Society,2009:97-99.
[9] HUANG G,CHENG C.3D Coordinate Identification of Object Using Binocular Vision System for Mobile Robot[C]//Automatic Control Conference.Nantou:IEEE,2014:91-96.
[10] 罗陆锋,邹湘军,叶敏,等.基于双目立体视觉的葡萄采摘防碰空间包围体求解与定位[J].农业工程学报,2016,32(8):41-47.
[11] 王磊,胡超,吴捷,等.基于多相机的人脸姿态识别[J].计算机应用,2010,30(12):3307-3310.
[12] 陈磊,高升,袁宝峰,等.基于多相机的空间机械臂视觉系统[J].航天返回与遥感,2014,35(3):35-42.
[13] 冯文灏.近景摄影测量[M].武汉:武汉大学出版社,2002.
[14] 万缨,韩毅,卢汉清.运动目标检测算法的探讨[J].计算机仿真,2006,23(10):221-226.
[15] 马吉甫,高雪峰.基于扫描线的连通域搜索算法[J].福建电脑,2006(12):121.