Remote Sensing Image Correction Algorithm To Study

The geometric correction is a very important step in the information processing of remote sensing. This geometric correction includes two parts. The first one is to establish some correspondent relationship between original space of aberrant picture and the space of standard picture. It makes the geometric orientation of pixels in remote sensing pictures, i.e. the correction of geographic latitude and longitude. The second part is to utilize this correspondent relationship to make all pixels in original space of aberrant picture appear on the correspondent positions on space of standard picture. Then it completes the calculation of brightness value of each image element. This thesis is to do careful research into the two parts of geometric correction.Firstly, the general remote sensing pictures contain thousands of ranges of pixels with a large mount of information. Therefore, it will cost a lot of time to do the traditional resampling, the thesis designs a resampling which based on the search in local area. This arithmetic sets a area which includes a point with no brightness value in middle, and does the search in this area. This method avoids do the search in whole image, and reduces lots of repeated job.Secondly, this thesis completes geometric correction in the district of second ring road of Chengdu by using rational function as correction model, and local searching to resampling. And then compare it with the traditional arithmetic, the results validates this arithmetic not only guarantees the picture's quality but also improves the speed of resampling of picture.Finally, as to correction of geometric coordinate, the polynomial model used frequently, the precision of this kind of arithmetic's correction has a close connection with control points, and it is hard to get control points, a differential geometric method to rectify digital images of the earth's surface proposed by Mahmut O. Karslioglu solves this matter. The thesis combines differential geometric method to rectify and local search to resampling fast, then designs a geometric correction based on differential geometric method and fast resampling, and do an emulation experiment by a man-made image, the result shows the validity of this arithmetic. In addition, the MODIS(Moderate Resolution Imaging Spectra radiometer)data can be got freely, and it includes geographical location information, so MODIS turns into the heated subject for research all over the world rapidly. The thesis does the preparatory research on the applying differential geometric method to rectify to MODIS Level 1A earth location, and expects it could improving the precision of MODIS Level 1A earth location.The arithmetic in this thesis is carried out by VC++.