The Research On Topographic Correction Of Remote Sensing Images Based On ETM+Data

Recently, the remote sensing technology has been more and more used in various fields.The remote sensing technique has been the key tool to acquire the earth’s surface information.In the analysis and applications of digital remote sensing images especially quantitativeremote sensing, the basis of quantitative calculation is the spectral signature of the surfaceobject. And the influence on spectral information of images which caused by the mountainousdistrict is the most obvious. Because of the impact of hypsography, there might be a greatdifference in effective illumination received by the earth’s surface element. Even to the samekind of target, the pixel in different gradient or slope will display different brightness value.On the other side, the surface features in different type may have similar brightness valueunder the influence of gradient and slope. This phenomenon of different object with the samespectra characteristics or different spectra characteristics with the same object disturbs thespectral signature information in images, which influence automatic classification accuracy ofremote sensing image data, and lead to the loss of accuracy of surface parameters’ quantitativeretrieval, with the result of hindering intensive quantitative application of remote sensing data.Topographic correction is one of the major tasks for remote sensing imagery correction, and itis also a necessary step to get the real reflectance from surface. We should only eliminate theinfluence of topography in the broadest sense by topographic correction to do the furtherprocessing of remote sensing precisely. Then we can extract the earth’s surface informationthat conform to the physical truth from the remote sensing images, and improve the accuracyof remote sensing monitoring in mountainous area.I process the geometry correction on the original image data with the purpose ofeliminating the geometric distortion at first in this paper. Then I process the atmosphericcorrection on the result obtained from last step by using the extension module named ACTORin ERDAS9.2, and get the result of the atmospheric correction. Finally, I use the5commontopographic correction models at home and abroad, which include cosine correction model, Ccorrection model, SCS correction model, SCS+C correction model and Minnaert correctionmodel, to process topography correction on the resulted image. The final result show that thebrightness contrast has been enhanced and the earth’s surface information in terrain shadowareas has been recovered in a large extent after the topographic correction. We can find outthat the Minnaert correction model comes to the best result, and the C correction model takessecond place. The results of cosine correction model and SCS correction model turn up to acertain extent of overcorrection.