基于主成分分析法的DEM粗差定位Detecting gross errors in DEM of regular data based on PCA
杨晓云,梁鑫,岑敏仪,顾利亚
摘要(Abstract)：
规则格网DEM数据结构简单,易于存储,因而针对于它设计的粗差探测算法成果丰富,较为常用的有基于坡度信息算法以及基于参数统计方法。本文在现有算法的基础上,将主成分分析法应用到DEM粗差定位中,充分考虑DEM数据空间相关的特性,使粗差检测更为准确可靠。本文采用实测的ZX铁路线DEM数据对该算法进行检验,从试验结果可以得知,新算法不仅适用于生产者,也可面向最终用户。
关键词(KeyWords)： 数字高程模型;粗差定位;主成分分析法
基金项目(Foundation): 国家自然科学基金资助项目(40271092);; 香港特别行政区研究基金委员会部分资助项目(编号:香港理工大学5068/99E)
作者(Author): 杨晓云,梁鑫,岑敏仪,顾利亚
参考文献(References)：
- [1]Hannah M J.Error detection and correction in Digital Terrain Models[J].Photogrammetric Engineering and Remote Sensing.1981,47:63-69.
- [2]Felicsimo A.Parametric statistical method for error de-tection in digital elevation models[J].ISPRS Journal of Photogrammetry and Remote Sensing.1994,49:29-33.
- [3]FERNAUME and Samson PJ.Use of cluster analysis to define periods of similar meteorology and precipitation chemistry in eastern North America[J].Part I:Trans-port Patterns.Journal of Applied Meteorology,1990,29.
- [4]Zhang You-shui,Feng Xue-zhi and Zhou Cheng-hu.Rela-tive radiometric correction for multitemporal TM image[J].Acta Geodaetica et Cartographica Sinica,2006,(2).
- [5]Wang Kun,Pan Ji-nong.A PCA based unsupervised a-nomaly detection approach[J].Journal of Information Engineering University,2004,3:57-59.
- [6]Richman M B.Rotation of principal components[J].Journal of Climatology,1986,6:293-335.
- [7]Lebart L,Morineau A and Tabard N.Techniques de la description statistique:Methodes et logiciels pour l’an-alyse des grands tableaux[M].Paris:Dunod,1997.
- [8]Hawkins D M.The detection of errors in multivariate data using Principal Components[J].Journal of the Ameri-can Statistical Association,1974,69:340-344.
- [9]Lpez C and Kaplan E.Principal component analysis ap-plied for outlier location and the missing value problem u-sing surface wind and pressure data[DB/OL].Internal report593.Available from http://www.ng.edu.uy/ncarlos/papers/rep93±5/viento.htm.1993:22.