关键词: 相位解缠, 非线性最小二乘, 地形坡度, 最小二乘模型, 梯度概率密度

Abstract:

Interferometric synthetic aperture radar (InSAR) phase unwrapping is one of the key technologies, which, uses the InSAR interference phase, to extract digital elevation model or accurate differential interferometry. However, affected by the porblem that the terrain is streep or the slope is larger, the unwrapping result is bad and causes error transmission in the differential wrapped phase information. In view of this problem, this paper considered to express the terrain slope in the interferogram as the partial phase frequency in range and azimuth direction. Using partial phase frequency to estimate the terrain slope and derivate the wrapped phase gradient probability density function (PG-PDF), the parameter model was used as the constraints of the nonlinear least squares phase unwrapping algorithm, in order to smooth the un-requirements unwrapped phase gradient. After unwrapping, the iterative solution which obtained from the results could eliminate the noise while reducing topographical factors under the condition of less sampling, at the same time the phase unwrapping results improved the accuracy of phase unwrapping. Finally, in the experiments that used the interferometric data obtained from ESA ENVISAT ASAR, it is verified that taking into account the terrain in the frequency domain method could effectively overcome the shortcomings of LS estimates for the phase gradient owed, and the algorithm in unwrapping could effectively give consideration to terrain factors, suppress error propagation, and have precision and adaptability to the terrain slope better than a direct weighted phase unwrapping algorithm.

Key words: topographic slopes, least squares model, gradient probability density, nonlinear least squares, phase unwrapping