The geometric correction and accuracy assessment based on Cartosat-1 satellite image

In the fast developing economic era, traditional topographic map updates can not keep up with the fast pace of the times, so taking use of remote sensing image data to make digital orthophoto map have been the most powerful means. In the article, it takes use of ERDAS LPS module to process Indian high-resolution satellite Cartosat-1´s (also known as IRS-P5) three-dimensional image extracting region DEM (digital elevation model). In support of DEM, we can produce the orthophotos, using rational function model. In order to compare the accuracy of orthophotos, which is made by different model, we also use the direct linear transformation model and polynomial model to calibrate the satellite image. we evaluate the positioning accuracy, according to the ground control points. Compare the accuracy of the different model and obtain the experimental results, it shows that under good control, rational function model´s location accuracy is higher than direct linear transformation model and polynomial model, especially in the undulating areas.