基于极大似然估计采样一致性准则的遥感影像配准参数解算方法研究Resolving method of remote sensing image registration coefficient based on Maximum Likelihood Estimation Sample Consensus
宋妍,田玉刚,贾小霞
摘要(Abstract):
本文提出运用最大似然采样一致性准则解算遥感影像配准系数的方法。该方法基于极大似然估计理论,首先对初始匹配点的坐标残差进行概率建模,计算概率模型成立时的似然函数值并选择似然函数值最大时的参数为正确结果,最终剔除错误点保留正确匹配点。该方法较之传统的最小二乘方法更为准确地计算配准系数,并可以解决随机采样一致性准则解算配准参数时,对阈值的依赖问题。试验证明,该方法可提高配准参数解算的稳健性和精度。
关键词(KeyWords): 极大似然估计采样一致性准则;遥感影像;配准;精度
基金项目(Foundation): 国家自然科学基金(NO.40801213);; 国家科技支撑“长三角地区自然灾害风险等级评估技术研究”(2008BAK50B07);; 中国地质大学(武汉)优秀青年教师资助计划(CUGQNL0932)
作者(Author): 宋妍,田玉刚,贾小霞
DOI: 10.16251/j.cnki.1009-2307.2011.01.076
参考文献(References):
- [1]Brown L G.A survey of Image Registration Techniques[J].ACM Computing Surveys,1992,24(4).
- [2]袁修孝,宋妍.基于边缘特征匹配的遥感影像变化检测预处理方法[J].武汉大学学报(信息科学版),2007,32(5).
- [3]TKim,Y J I.Automatic Satellite Image Registration byCombination of Matching and Random Sample Consensus[J].IEEE Transaction of Geoscience and Remote Sens-ing,2005,41(5).
- [4]钟家强.基于多时相遥感图像的变化检测[D].长沙:国防科技大学,2005.
- [5]文贡坚.从新卫星遥感影像中自动发现变化区域〔R〕.武汉:武汉大学博士后研究工作报告,2003.
- [6]Fischler Martin A.Random Sample Consensus:A para-digm for Model Fitting with applications to Image analysisand automated cartography[J].Graphics and ImageProcessing,1981,24(6).
- [7]Chum O.Two-View Geometry Estimation by RandomSample and Consensus,in Department of CyberneticsFaculty of Electrical Engineering[D].Czech Techni-cal University in Prague,2005.
- [8]Torr PH S,A Z MLESAC.A new robust estimator withapplication to estimating image geometry[J].Comput-er Vision and Image Understanding,2000,78(1).
- [9]Khotanzad A,H Y H.Invariant image recognition byZernike moments[J].IEEE Transactions on PatternAnalysis,Machine Intelligence,1990,12(5).
- [10]徐健斌,洪文,吴一戎.一种基于Zernike矩和稳态遗传算法的遥感图像匹配方法[J].电子与信息学报,2005,27(6).
- [11]Habib A,Al Ruzouq R.Semi-automatic Registration ofMulti-source Satellite Imagery with Varying GeometricResolution[J].Photogrammetric Engineering&Re-mote Sensing,2005,71(3).
- [12]Rousseeuw PJ,Leroy A M.Robust Regression andOutlier Detection[M].New York:WileyPress,1987.
- [13]陈希孺.数理统计教程[M].北京:科学出版社:1984:55-91.