Research On High Precision Processing Methods For High Resolution Optical Satellite's Jitter Measurement Data

High-resolution optical remote sensing satellite(HRORSS)has become an important part of the earth observation technology.The high precision geometric processing quality of image is the key and foundation for the following processing and application.Low frequency and accuracy of the jitter measurement system,and the satellite platform jitter have become an important bottleneck of the HRORSS image processing.The satellite platform jitter mainly refers to the smaller amplitude and higher frequency jitter response,which are caused by external factors,such as the attitude control of the satellite platform,the adjustment of the solar array pointing,liquid fuel sloshing and so on.The satellite platform jitter leads to unsteady pushbroom imaging of linear camera,images have the certain distortion,affecting the internal geometric imaging quality,and with the future of China's 0.3 or even higher resolution optical remote sensing satellite launch,the impact will reach more than ten pixels,unable to meet the high precision remote sensing applications such as global mapping and change detection.To be aimed at the influence of the satellite platform jitter on the imaging quality,on the one hand,we use the physical vibration isolation measures to suppress;on the other hand,we mainly use the parallax imaging principle to detect and compensate the jitter image.The above methods have some limitations,belong to the passive processing method,and the effectiveness is limited.In recent years,with the successfully development of new type of jitter measurement sensor such as linear accelerometer,laser gyro(high frequency angular displacement sensor)and so on,which have been equipped with a new generation of submeter optical remote sensing satellite,realizes the direct detection of satellite platform jitter information,and provides new ideas and methods for on-ground jitter detection,modeling and compensation of HRORSS image.To be aimed at the above problems and challenges,this paper has done research on the high precision processing and application of HRORSS jitter measurement data.On the basis of doing research on the error characteristics analysis and calibration of optical satellite jitter measurement system,we proposed the method of high accuracy information fusion processing of optical satellite's multi-source jitter measurement data and optical distortion image reconstruction and restoration based on jitter measurement,and used the data of YG-XX,GF-XX satellites for the experimental analysis and validation.The contents and innovations are listed as follows:(1)The method of jitter measurement system error characteristics analysis and calibration for optical image geometric processing was proposed,which layed the theoretical foundation for high accuracy information fusion processing of multi-source jitter measurement data.The measurement principle and error sources of the attitude sensor(star sensor,three floating gyroscope,fiber optic gyro,laser gyro and linear accelerometer)were systematically studied and discussed,then the quantitative relation between the error terms of the jitter measurement system and the geometric processing of the image was analyzed,and the error measurement model was constructed from the perspective of HRORSS image geometric processing.Based on the above research foundation,we proposed the method of detection,analysis and calibration for jitter measurement system's installation error,internal error characteristic parameter and time-varying system error,which layed the theoretical foundation for subsequent high accuracy information fusion processing of multi-source jitter measurement data.(2)The method of high accuracy information fusion processing of optical satellite multi-source jitter measurement data was proposed,which provided a new way for the accurate detection of the platform jitter signal.To be aimed at the characteristics of internal attitude sensor diversity in jitter measurement system and the complexity of error measurement model,we did reasonable choice,approximation and optimization of system error parameters from the perspective of image geometric processing,proposed generalized high precision information fusion model for absolute attitude measurement subsystem and relative attitude measurement subsystem to achieve accurate estimation and detection of platform jitter information.It included the high precision absolute attitude calculation and construction of generalized measurement model based on absolute attitude measurement subsystem,construction of generalized error measurement model and state model based on relative attitude measurement subsystem,construction of the information fusion model based on the bidirectional filter,overall weighted adjustment and the optimal estimation of attitude parameters.(3)The method of optical distortion image reconstruction and restoration based on platform jitter measurement was proposed,which could effectively improve the internal geometric quality of optical remote sensing image.Due to the inconsistency of the optical remote sensing satellite platform and linear array camera's jitter signal,the platform jitter signal can not be directly used for on-ground detection and compensation of optical distortion image.In order to solve the key scientific problems,we proposed using the geometric calibration field,the high precision dense matching method and the rigorous geometric imaging model of probe element's direction angle to realize the calculation of linear array camera's real jitter signal,then did spectral decomposition of the platform and the linear array camera's jitter signal with fourier transform and constructed the jitter transfer model.Finally,we used the method of virtual steady reimaging to realize the reconstruction and restoration of the optical distortion image,which eliminated the influence of satellite platform jitter on the image geometric quality of optical image.In the view of the research contents above,the error characteristics analysis and calibration of the jitter measurement system,the information fusion processing and the optical distortion image reconstruction and restoration method were verified by using the real data of YG-XX,GF-XX satellites,and the experimental results showed that the method proposed in this paper could accurately identify,calibrate and model the installation error,internal error characteristic parameters and time-varying system error of the jitter measurement system,realize the high accuracy estimation of the platform and linear array camera jitter information,and the construction of tremor transfer model,which effectively improved the external and internal geometric quality of HRORSS jitter distortion image.