In-Orbit Autonomous Imaging Technology Of Space TDI CCD Camera

The goal of in-orbit autonomous imaging technology is that spaceborne camera could auto adjusting imaging parameters and image recover in-orbit, in order to improve the autonomous imaging capacity. The feature information of camera’s performance extracted from images, add to the camera’s imaging model, the camera’s imaging parameters could auto adjust, together with in-orbit image recover, the camera could achieve closed loop control as "images to parameters to images", for the camera, the dependence is reduced greatly to attitude, orbit information and imaging parameters.In this desertion ,the Diffusion Radius (DR) of the system Point Spread Function (PSF) is used to measure the quality of imaging, established imaging electrical model to connect the DR and the imaging parameters, Step Response Method (SRM), Power Spectrum Cut-off frequency Method (PSCM) and Wavelet High-frequency Energy Method (WHEM) were proposed to constitute a Triple Modular Redundancy (TMR), which extracted DR together. using DR and imaging electrical model ,the rough values of imaging parameters could be calculated, the precise values could be searched in the neighborhood; in order reduce the dependence to the focusing mechanism and yaw mechanics, and to overcome the imaging deteriorating when the orbit attenuated, in-orbit image recover algorithm was proposed.In order to establish the imaging electrical model, imaging performance were divided into high-frequency character (HFC) and low-frequency character (LFC), HFC was measured by DR, and LFC was measured by the Mean of Gray Value (MGV); based on the principle and existing research results, proposed the PSF of defocus process, yaw process and image motion mismatch process, and established imaging electrical model to connect the DR and the imaging parameters, proposed the Optimization function of TDI-stage and gain to achieve highest Signal Noise Rate(SNR).In order to extract image feature as feedback signal, the character value extracted from image need reflect the imaging performance objectively, so that the camera could auto optimization the amount of defocus, yaw angle, TDI CCD line transfer cycle (LTC), TDI stage and gain. Based on imaging electrical model, extracting Low Frequency Character Value (LFCV), Channel Equivalent Character Value (CECV), and High Frequency Character Value (HFCV) of X and Y direction, to complete all the imaging parameter adaptive adjusting. TDI CCD imaging is line scanning way, imaging target(Earth surface feature) changing over time, feature extracting algorithm must be effectively inhibited to the surface feature, this desertion using multi-sample parallel processing to eliminate the impact of the random character; TMR concept was introduced in image processing, HFCV was extracted parallel by SRM, PSCM and WHEM, so that the rang and precision is both better than single method, based on which imaging electrical model estimated the rough value of imaging parameters.In order to achieve the adaptive adjustment of imaging parameters, this desertion presents a Four-Step Fast Algorithm (FSFA) combination of estimation and the search, the first step, according to DR to calculate the rough value of imaging parameters, the second step, to determine the direction of transfer by trying adjust, the third step, to adjust to the rough value, the fourth step, to search precise values. According to the parameter adjusting flow, complete the process of focusing, yaw adjusting, image motion matching and the channel calibrated, in the adaptive imaging phase, based on feedback signal, adaptively adjusting LTC, yaw angle, gain and TDI stage, So the camera could maintain optimal imaging condition.In order to reduce the dependence to the focusing and yaw mechanism, and to overcome the imaging deteriorating as the orbit attenuated, image recover Algorithms for defocus, yaw and image motion mismatch were proposed. According to TDI CCD imaging principle, established TDI CCD Energy Transfer Model (TDI-ETM), which established a relationship between system point spread function and image surface features contribute rate, combination imaging electrical model, proposed defocus, yaw, and image motion mismatch image restoration algorithm, so the camera can still get higher quality images.Imaging experiments show that this method of image feature extraction could adapt to a wide range and high precision parameter adjusting , has a good adaptability to different objects ; imaging parameters could rapid changes in adaptive according to imaging conditions; image feature extracting, parameter adjusting and image restoration formed the basis of space camera of intelligent imaging technology, the intelligent imaging were verified through imaging experiments, the effect is consistent with the simulation results.