3D Imaging Simulation And Elevation Inversion For Sea Surface And The Error Analysis And Correction

The ocean covers 71% of the earth's surface.It has been paid more and more attention all over the world with the development of economy and society,especially the improvement of science and technology.Remote sensing has become an important means of ocean monitoring because of its advantages of large coverage,high observation frequency,all day,all-weather and so on.One important task of ocean remote sensing is obtaining the sea surface height(SSH)data,which is also an important information support for mastering the state of marine strategic environment and marine disaster early warning.At present,the SSH measurement mainly uses traditional radar altimeter.It can complete one-dimensional height measurement along the trajectory of the satellite.Due to the large orbital interval,its resolution is low and it can not implement the sub-mesoscale area measurement of SSH.Therefore,the three-dimensionl(3D)imaging altimeter has been studied at home and abroad.It has the advantage of imaging,large coverage area,high resolution,good real-time performance.Also,it can meet the sub mesoscale ocean phenomenon observation,and can observe both the ocean and land.Taking 3D imaging altimeter as the application object,the algorithms of sea surface imaging and elevation inversion are studied by using computer simulation technology.The main error factors of 3D imaging altimeter height measurement are analyzed,and the corresponding correction methods are designed.It provides algorithm support for improving the accuracy of SSH.The main accomplishments are as follows:1.The sea surface imaging of 3D imaging altimeter is studied.One sea surface imaging simulation algorithm is proposed according to the main characteristics and parameters of 3D imaging altimeter.First,the two-scale model is used to simulate the sea surface.Then,it is divided into small triangular facets.Second,the backscattering cross sections of these small facets are calculated via application of quasi-mirror scattering mechanism and its Kirchhoff approximate solution,and the backscattering coefficient of the simulated region derived via coherent superposition.Third,system parameters are set consistent with the basic principle of 3D imaging altimeter to establist the echo signal model.The simulation images are then derived using the Range-Doppler(RD)and Back-Projection(BP)algorithms.When constructing the echo signal model,the equivalent phase center processing was performed for the dual-antenna interference mode of the 3D imaging altimeter.The results of imaging simulation also can provide images data for elevation inversion and error analysis and correction.2.The images data processing of 3D imaging altimeter is studied.The designed SSH elevation inversion algorithm includes five basic steps: images registration,removing flat earth effect,filtering of interference phases,phase unwrapping and converting phase to elevation.In image registration,an algorithm combining enhanced Scale Invariant Feature Transform(SIFT)features with correlation coefficients is proposed.It achieved sub-pixel level accurate images registration.Its registration accuracy and scope of application are better than using the SIFT algorithm and the correlation coefficient method alone.In removing flat earth effect,the method of moving the main spectrum center to 0 in the frequency domain is adopted.The algorithm combinating median pre-filtering and multi-directional fusion linear filtering is used in the interferometric phase filtering.In the phase unwrapping,the branch cuts are generated using the Jonker-Volgenant-Castanon(JVC)global optimal linear allocation algorithm to improve the classic Goldstein algorithm.It shortens the total length of the branch cuts.And,to a certain extent,it can avoid the“island” problem which cannot be unwound locally.The phase unwrapping rate and accuracy are improved.Finally,the The relative SSH values are calculated according to the principle of interferometry and the observation geometry of 3D imaging altimeter.The simulation images of the sea surface is used for elevation inversion.The feasibility of the sea surface imaging simulation algorithm is verified by comparision.3.The factors affecting the accuracy of SSH measurement for 3D imaging altimeter are studied.The errors of system parameters,background ionospheric,background atmospheric and the phase pollution of ionospheric scintillation and atmospheric turbulence are set.Based on the imaging simulation and elevation inversion,the influence of various errors on the SSH measurement accuracy is analyzed by comparing the original simulated SSH with the inversion results.The results show that the most significant errors come from the ionospheric scintillation and atmospheric turbulence.To correct the phase errors,two algorithms are proposed.one is improving the combination of Phase Gradient Autofocus(PGA)and Map Drift(MD)algorithms by using the methods of adaptive point selection based on neighborhood pixel mean value and adaptive estimation for window width based on waveform contour extraction.The other is improved the method based on the minimum entropy criterion by through adopting the Fruit Fly Optimization Algorithm(FOA)to search the best phase errors correction values in the range compression phase history domain.Simulation results show that both the two proposed algorithms can get obvious correction effect.But,the method based on FOA and minimum entropy criterion is more effective.4.Considering the influence of system parameter errors and atmospheric disturbance on sea surface elevation measurement by 3D imaging altimeter,the synthetical experiment is developed with the proposed methods of the above three parts work.It includes sea surface imaging simulation,errors correction and elevation inversion.Its results prove that the whole research work and the proposed algorithms are reasonable and effective.