Rectification For High Resolution Remote Sensing Image Under Lack Of Ground Control Points

Earth observation by satellites is the important method of acquisitioning the spatial information on terrestrial space, and is producing important result on national economic construction and national defense construction. For geometrical measuring, reciprocal comparison, image compound analysis and etc., the remote sensing images always need to be expressed at a certain projection reference system and eliminate geometric distortion of imagery.How to project remote sensing images onto a certain reference system and keep geometrical position, shape, dimension and orientation of objects unchanged, namely geometrical rectification of remote sensing images, is an essential process in remote sensing image's processing and application. From of old, the main method of geometrical rectification of satellite images is polynomial fitting that depends on the amount and well distribution of ground control points (GCPs). But acquisition of GCPs in distressed area such as desert, ocean, borders, oversea, and so on is difficult or even impossible. So research on rectification of remote sensing image under lack of GCPs is becoming an urgent task. Furthermore, the IPS (image processing system) of the domestic satellite relies on importation. In order to get good results, we have no choice but provide foreign experts with core parameter of satellite. It brings hidden danger to national security. The key to IPS is systematic geometrical rectification, namely geometrical rectification without GCPs, so research on rectification of remote sensing image under lack of GCPs aim at the domestic satellite is live on development our IPS.This paper focuses on the theory and method of rectification of remote sensing image under lack of GCPs. Its idea can be expressed as the following:The study on mathematical model is the key of the rectification of remote sensing image. After delving into the pay-off on geometrical rectification, this paper set up rigorous sensor model of high resolution remote sensing image which considers the rigorous transformation between image coordinate and the project coordinate and the establishment of orbit model and attitude.The rigorous sensor model is the ideal sensor model which does not consider the error about the CCD installation. This paper modifies the computational method of set over matrix which can be used to redress the bias error based on the analyzing theoretical precision of remote sensing image.According to the orbital dynamical model and the kinematics model of satellite attitude and the side look angle stability, mis paper developed the extrapolative location model.Studying on the computational method of the RPC parameters based on rigorous sensor model, the algorithm computing the RPC parameters which take global DEM in consider and does not need init parameter is bung up.Studying on the GCP measure, this paper modifies the hill climbing method in order to assure the best matching of GCP image and uses elliptic paraboloid fitting to improve GCP matching precision.Based on the study of the algorithms, the pilot system on rectification of remote sensing image under lack of GCPs is developed, which can be used to rectify the SPOT-5 image, QUICKBIRD image, IKONOS image and CBERS2 image under lack of GCPs, and to locate CBERS2 image without any GCPs by extrapolative location model.