Please wait a minute...
img

官方微信

遥感技术与应用  2013, Vol. 28 Issue (4): 604-609    DOI: 10.11873/j.issn.1004-0323.2013.4.604
图像与数据处理     
基于SIFT和TPS算法的四旋翼无人机图像自动配准
陈本清,杨燕明,郑凌虹,文洪涛
(国家海洋局第三海洋研究所,福建 厦门 361005)
Automatic Registration of Quad-Rotor UAV Imagery based on SIFT and TPS Algorithm
Chen Benqing,Yang Yanming,Zheng Linghong,Wen Hongtao
(Third Institute of Oceanography,State Oceanic Administration,Xiamen 361005,China)
 全文: PDF(2753 KB)  
摘要:

针对四旋翼无人机图像姿态倾角大、图像变形明显等特点,采用尺度不变特征变换(SIFT)算法和薄板样条模型(TPS)对四旋翼无人机图像进行特征点匹配和配准实验研究,从拼接图像的目视效果和配准均方差方面比较分析了TPS模型与常用的仿射变换及多项式变换模型的图像配准效果。结果表明:在SIFT算法精确的同名点匹配下,TPS变换模型能够兼顾四旋翼无人机图像的整体刚性变形及局部的非刚性变形,无论是目视效果还是均方差定量分析,TPS变换的图像配准精度最高\,效果最好,能够满足四旋翼无人机图像的快速配准、拼接要求。

关键词: SIFT特征匹配图像配准TPS模型四旋翼无人机    
Abstract:

As an important complementarity of remote sensing technique,Unmanned Aerial Vehicle (UAV) is attractive to various applications such as target searching,island management,disaster monitoring and low-altitude photogrammetry.The quad-rotor UAV,Compared to fixed-wing UAV,has the advantages of Vertical-Taking-of-Landing (VTOL) and low-altitude flexible flight.However,the quad-rotor UAV has small-volume and light-weight and is more easily influenced by the wind,the acquired imagery maybe have bigger tile angle and more obvious geometry distortion will would result in more difficulty on the image feature matching and  image mosaic. For this question,we apply the Scale Invariant Feature Transformation (SIFT) to image feature matching and Thin Plate Spline (TPS) transformation to automatic registration of the micro quad-rotor UAV imageries in this paper.The registration imagery based on TPS transformation is then compared to that based on the affine transformation and the polynomial transformation by evaluating the visual effect of the mosaic imagery and Root Mean Square (RMS) statistic,which shows that after precise SIFT feature matching,the registration RMS accuracy and the visual effect of mosaic imagery of TPS transformation are best,and satisfies the demand of rapid registration and mosaic of micro quad-rotor UVA imagery,since TPS transformation considered both rigid transformation and partial nonlinear distortion of the micro quad-rotor UAV imagery.

Key words:      SIFT feature matching    Image registration    TPS transformation    Quad-Rotor UAV
收稿日期: 2012-05-22 出版日期: 2013-08-14
:  TP 75  
基金资助:

国家海洋局第三海洋研究所基本科研业务费专项资金(2010019),海洋公益性行业科研专项经费资助(201005010),国家海洋局第三海洋研究所基本科研业务费专项资金(2010013).

作者简介: 陈本清(1977-),男,福建长乐人,副研究员,主要从事环境遥感及无人机遥感应用研究。E-mail:benqing_chen@163.com。
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
陈本清
杨燕明
郑凌虹
文洪涛

引用本文:

陈本清,杨燕明,郑凌虹,文洪涛. 基于SIFT和TPS算法的四旋翼无人机图像自动配准[J]. 遥感技术与应用, 2013, 28(4): 604-609.

Chen Benqing,Yang Yanming,Zheng Linghong,Wen Hongtao. Automatic Registration of Quad-Rotor UAV Imagery based on SIFT and TPS Algorithm. Remote Sensing Technology and Application, 2013, 28(4): 604-609.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2013.4.604        http://www.rsta.ac.cn/CN/Y2013/V28/I4/604

[1]Ma Zezhong,Wang Fuhai,Liu Zhihua,et al.Research of Unmanned Aerial Vehicle Remote Sensing Technique Applied to Monitor Landslide and Barrier Lake Hazard in Chengkou County of Chongqing[J].Journal of Soil and Water Conservation.2011,25(1):253-256.[马泽忠,王福海,刘智华,等.低空无人飞行器遥感技术在重庆城口滑坡堰塞湖灾害监测中的应用研究[J].水土保持学报,2011,25(1):253-256.]

[2]Gao Jiaojiao.Application of High Precision UAV Remote Sen-sing in Geological Disaster Investigation[D].Beijing :Beijing Jiaotong University,2010.[高姣姣.高精度无人机遥感地质灾害调查应用研究[D].北京:北京交通大学,2010.]

[3]Han Jie,Wang Zheng.On the Key Technique of Fast Monitoring System for Resources Unmanned Plane Remote Sensing[J].Bulletin of Surveying and Mapping,2008,(2):4-7.[韩杰,王争.无人机遥感国土资源快速监察系统关键技术研究[J].测绘通报,2008,(2):4-7.]

[4]Yang Yanming,Zheng Linghong,Wen Hongtao,et al.Application of UAV Remote Sensing Technology in the Management of Island[J].Ocean Development and Management,2011,(1):6-10.[杨燕明,郑凌虹,文洪涛,等.无人机遥感技术在海岛管理中的应用研究[J].海洋开发与管理,2011,(1):6-10.]

[5]Delacourt C,Allemand P,Jaud M,et al.DRELIO:An Unmanned Helicopter for Imaging Coastal Areas[J].Journal of Coastal Research,SI 56:1489-1493.

[6]Zhang Yongjun.Geometric Processing of Low Altitude Remote Sensing Images Captured by Unmanned Airship[J].Geomatics and Information Science of Wuhan University,2009,34(3):284-288.[张永军.无人驾驶飞艇低空遥感影像的几何处理[J].武汉大学学报:信息科学版,2009,34(3):284-288.]

[7]He Jing,Li Yongshu.A Mosaic Method in Unmanned Aerial Vehicle Images based on Feature Points and Optimal Path[J].Remote Sensing Technology and Application,2012,27(2):173-176.[何敬,李永树.基于特征点和最优路径的无人机影像拼接方法[J].遥感技术与应用,2012,27(2):173-176.]

[8]Chen Yu.The Unmanned Aerial Vehicles Remote Sensing Image Registration based on SIFT Algorithm[D].Changsha:Central South University,2009.[陈裕.基于SIFT算法的无人机遥感图像配准[D].长沙:中南大学,2009.]

[9]Zhang Ruijuan,Zhang Jianqi,Yang Cui.Image Registration approach based on SURF[J].Infrared and Laser Engineering,2009,38(1):160-165.[张锐娟,张建奇,杨翠.基于SURF的图像配准方法研究[J].红外与激光工程,2009,38(1):160-165.]

[10]Lowe D G.Distinctive Image Features from Scale-invariant Keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.

[11]Mikolajczyk K,Schmid C.A Performance Evaluation of Local Descriptors[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(10):1615-1630.

[12]Li Xiaoming.SIFT based Automatic Registration of Remotely Sensed Imagery[J].Journal of Remote Sensing,2006,10(6):885-892.[李晓明.基于SIFT特征的遥感影响自动配准[J].遥感学报,2006,10(6):885-892.] 

[13]Cheng Yan,Zhou Yan,Lin Hongtao,et al.Automatic Registration and Mosaic of Remote Sensed Imagery based on SIFT Feature[J].Remote Sensing Technology and Application,2008,23(6):721-728.[程焱,周焰,林洪涛等.基于SIFT特征遥感影像自动配准与拼接[J].遥感技术与应用,2008,23(6):721-728.]

[14]Qian D,Nareenart R,Adnan O.et al.Automatic Registration and Mosaicking for Airborne Multispectral Image Sequences[J].Photogrammetric Engineering & Remote Sensing,2008,74(2):169-181.

[15]Goshtasby A.Registration of Images with Geometric Distortions[J].IEEE Transactions on Geoscience and Remote Sensing,1988,26(1):60-64.

[16]Zagorchev L,Goshtasby A.A Comparative Study of Transformation Functions for Nonrigid Image Registration[J].IEEE Transactions on Image Processing,2006,15(3):529-538.

[17]Li Yuling.Sutdy of Mosaicing Method in High Resolution Remote Sensing Images without Geogarphy Coordinate[D].Beijing:Institute of Remote Sensing Application,Chinese Academy of Sciences,2005.[李豫玲.无地理坐标高分辨率遥感影像镶嵌方法的研究[D].北京:中国科学院遥感应用研究所,2005.]

[1] 侯鹏洋,季艳,高峰,胡蕾. 一种基于SIFT特征的快速逐层遥感图像配准方法[J]. 遥感技术与应用, 2014, 29(5): 873-877.
[2] 汤益先,赫永杰,张红,王超. 基于FFT平移与相关分析的多尺度星载SAR自动配准[J]. 遥感技术与应用, 2013, 28(4): 618-626.
[3] 欧阳能钧,李伟彤,韦蔚,潘晴. 基于SIFT与Contourlet变换的高分辨遥感图像配准[J]. 遥感技术与应用, 2013, 28(1): 58-64.
[4] 周 刚, 许德伟, 杨燕明, 刘贞文, 傅世锋. 基于IDL语言与控制点影像数据库的图像自动配准技术研究[J]. 遥感技术与应用, 2010, 25(5): 627-631.
[5] 程焱,周焰,林洪涛,潘恒辉. 基于SIFT特征遥感影像自动配准与拼接[J]. 遥感技术与应用, 2008, 23(6): 721-728.
[6] 刘子龙,董臻,蔡斌,孙造宇. 星载InSAR非相干图像对精确配准[J]. 遥感技术与应用, 2008, 23(4): 446-450.
[7] 刘 广,郭华东,范景辉. 基于外部DEM的InSAR图像配准方法研究[J]. 遥感技术与应用, 2008, 23(1): 72-76.
[8] 霍春雷,周志鑫,刘青山,卢汉清. 基于SIFT特征和广义紧互对原型对距离的遥感图像配准方法[J]. 遥感技术与应用, 2007, 22(4): 524-530.
[9] 马广彬,章文毅,刘定生,陈 甫. 一种新的卫星多光谱图像自动配准算法[J]. 遥感技术与应用, 2007, 22(1): 95-100.
[10] 梁 勇, 祝明波, 杨汝良. 一种基于Hausdorff 度量的多传感器图像配准方法[J]. 遥感技术与应用, 2006, 21(5): 473-476.
[11] 罗 欣, 郭 雷, 杨诸胜. 基于互信息和随机优化的超光谱遥感图像配准[J]. 遥感技术与应用, 2006, 21(1): 61-66.
[12] 邓 鹏,王宏琦. 结合边缘与灰度信息的SAR图像配准算法[J]. 遥感技术与应用, 2003, 18(3): 159-163.
[13] 周红妹,杨星卫,楼 萌. 气象卫星小范围区域遥感图像配准方法的探讨[J]. 遥感技术与应用, 1994, 9(1): 52-55.