The Research And Application On HJ-1On-Orbit Satellites Remote Sensing Images MTF Calculation Based On The Knife Edge Method

MTF, the modulation transfer function, is a comprehensive evaluation index of optical system imaging quality. MTF of Imaging system affect the imaging quality directly.when the MTF is higher, the edge texture details of the obtained remote sensing image are more clear, on the contrary, the edge texture details of the obtained remote sensing image are more blurred. For the orbiting satellite remote sensor, monitoring the dynamic change situation of MTF can provide a reference basis for on-orbit focusing, monitoring the running status of satellite, forecasting the life of on-orbit satellite remote sensor, restoring remote sensing image, and designing for subsequent satellite, etc. As domestic HJ-1(environmental satellite no.1) series of satellites have been put to use, the remote sensor performance of on-orbit MTF monitoring problems have also obtained more and more attention. In this paper, the main work is as follows:1. A algorithm based on edge for calculating the MTF of HJ-1satellite remote sensing image is put forward. First of all, starting from the basic principle of MTF, the data sampling, interpolation and fitting method have carried on the detailed introduction. According to the advantages and disadvantages of different interpolation methods, we chose the Newton interpolation method due to its characteristic of inherited and facilitate theoretical analysis and so on. What’s more, we propose a Newton edge fitting method by programming improvement. The method can reduce the complexity of computation time. For interpolation of n points, the complexity of computation time is from O(n3) to O(n2). Also it saves the storage unit. Secondly, in the process of computing the MTF of on-orbit satellite, the expert reference module is increased. Judging by edge uniformity and linearity,we can know that the edge selection and edge fitting are in line with the prediction of experts or not, so we can judge the accuracy of the edge selection and the the edge fitting method.2. This paper proposes a new method for restoring satellite remote sensing image based on the theory of the MTF. Due to the quality of satellite remote sensing image is falling with the MTF falls, there is a corresponding relationship for both of them directly. If the MTF decline curve of the imaging system is obtained, and the MTF and the corresponding point of the image quality is transformed Inversely, we can get back of remote sensing images. By texture analysis and gray analysis:the average, contrast, entropy and variance of remote sensing image after recovery are increased significantly. For the performance of images, the image visual effect is better significantly, the edge feature and city roads become more clear, information of image is change quantitatively and image grey level is more abundant.