The Key Technology Researching Of Satellite Linear Array Image

Every line array image of the push-broom sensor of linear CCD is center projection, and line scan image is the parallel projection. Each scan line image has its exterior orientation elements, that is to say, the exterior orientation of every line image is difficult. Due to the flight direction between array image distortion,it is difficult to establish strict nuclear line model. As a result, many scholars have conducted research for linear array image.In this paper, digital image processing method has the main technology : Image preprocessing image feature extraction stereo images like matching access point to the same name,Is the next line array imaging image point coordinate data for the study on dense point cloud. There are dense point cloud based on rational function model is studied。The main content of this paper is as follows:(1)Because of the reason of the satellite attitude and remote sensors,result to resources 3 original image data has some drawbacks. Such as geometric distortion within the image,including the length of the image internal deformation and Angle。So before the feature extraction, we need to image preprocessing.(2)The paper adopts the method of multistage image matching。First pyramid hierarchical method is used to analyze the three linear array image hierarchical processing,and then to feature extraction in each layer,the step by step for the extraction of feature points. And use the parallax and Markov distance constraints for the feature point matching by mistake. Then the sub pixel level matching,this paper uses Least Squares Image Matching, fine matching.(3)Matching point data obtained by the study of rational function model,to research by three linear array image matching points with the same study the corresponding three-dimensional coordinates of the model.