Research On The Technique Of Micro-pixel Centroiding

In recent years,the search for extraterrestrial life and exoplanet systems has become a hot topic for human space exploration.Especially the detecting achievement of the habitable exoplanets in the circumstellar habitable zone of the host star which have the similar mass with earth is the ultimate goal of human exploration.In order to meet the needs of application of astrometry method to detect the habitable planets outside the system,the centroid positioning accuracy of the star must reach 10 micropixels(1 micro arcsecond).Therefore,the technique of micro-pixel centroiding is one of the key techniques to applying astrometry method to detect habitable exoplanets.In this thesis,the research on the technique of micro-pixel centroiding is carried out.The specific research contents and main results are as follows:Firstly,the micro-pixel centroid estimation algorithm and the technique of calibrating pixel response function of the image sensor in frequency domain are systematically studied.The effects of rolling shutter and interference fringe instability on image sensor calibration are analyzed and the corresponding solutions are given.Considering the influence of the focal plane assembly,the micro-pixel centroid estimation algorithm and the technique of calibrating pixel response function of the image sensor in frequency domain are improved respectively.Regarding to the technique of calibrating pixel response function in frequency domain cannot calibrate pixel response function in spatial domain,a new calibration method is proposed to overcome this defect.Secondly,the accuracy of micro-pixel centroid estimation algorithm and the influence of various error sources are studied through simulation.The results show that in order to achieve the micro-pixel level accuracy,the calculation window must cover the fifth-order Airy ring of the optical system point spread function.After the pixel response is calibrated in frequency domain,the centroid positioning accuracy can be improved from 0.001 pixel to 5 micro-pixels.Thirdly,a micro-pixel centroiding testbed to verify the technique is designed.The theoretical accuracy of the pixel position offset of the testbed is 25.7 micropixels.The preliminary test was carried out by using the testbed.The results show that after the calibration of the flat field,the centroid measurement precision of the micro-pixel centroid estimation algorithm is 80 micro-pixels in the integration time of 200 s,and the accuracy reaches to 0.0024 pixels.Finally,some explorations have been made on how to apply the technique of micro-pixel centroiding to star sensors.The influence of pixel position offsets on centroid position accuracy of star sensors is analyzed.The results show that when the centroid position accuracy is higher than 0.01 pixels,it is necessary to calibrate the pixel position offsets.Then the micro-pixel centroid localization algorithm is improved,a star sensor centroid algorithm based on star image resampling is proposed.The experimental results show that the systematic error of the algorithm is 0.008 pixels.It concludes that the proposed algorithm has high precision and is not affected by the aberrations of the optical system,it is an effective method to improve the accuracy of star sensor.The innovation of the thesis is mainly reflected in the following four aspects:1.Aiming at the intensity and the velocity of interference fringe instability phenomenon found during the actual interference fringe data processing,a method using an iterative process to get phases of different pixels is proposed,which can accurately and quickly analyze unstable interference fringe data and get accurate calibration result of pixel response functions.2.For large scale focal plane assembly,an algorithm for reconstructing the effective point spread function and resampling on different image sensors is proposed,which effectively solves the problem of high-precision measurement of the distance between stars on different image sensors with a certain spinning angle.3.Regarding to the technique of calibrating pixel response function in frequency domain cannot measure pixel response function in spatial domain,a new calibration method is proposed to overcome this defect.This method combines the frequency domain calibration method and direct measurement method.It overcomes the limitations of direct measurement method and frequency domain calibration method,the pixel response function in spatial domain can be obtained.This method also has the advantages of fast speed and strong applicability.4.The application of micro-pixel centroid positioning technology in highprecision star sensors is explored.The influence of pixel offsets on the centroid position accuracy of star sensors is analyzed.A new centroid algorithm of the star sensor based on star image resampling is proposed.These results provide a new way to develop highprecision star sensors without increasing hardware resources.