一种基于结构函数的干涉图筛选方法

An interferogram selection method based on structure function

针对目前时序合成孔径雷达干涉测量(InSAR)技术中最常用的时空基线硬阈值法存在的干涉图筛选盲目性问题,该文通过引入结构函数,构建了两个基于缠绕干涉相位结构函数的干涉图质量判断指标,用于干涉图筛选,其相较于传统的时空基线硬阈值法能选出更多高质量的干涉图。分别利用结构函数法和时空基线硬阈值法筛选出相同数目干涉图,得到两种地表形变监测结果,并利用水准数据进行精度验证。结果表明:结构函数法得到的形变速率精度更高,其与水准观测值差值的标准差为5.58 mm/a,形变精度相较于时空基线硬阈值法提高了35.57%,结果验证了结构函数方法在干涉图筛选中的有效性。

Aiming at the blind problem of interferograms screening existing in the temporal-spatial baseline thresholding method most commonly used in time series InSAR(TS-InSAR)processing,by introducing structural function,two parameters built upon the structure function(SF)of the wrapped interferogram phase were employed to evaluate the quality of the interferogram,which was used for interferograms selecting,and could select more high-quality interferograms comparing with the traditional temporal-spatial baseline thresholding method.The same amount of interferograms were selected with the SF parameters method,and the temporal-spatial baseline thresholding method,respectively,two kinds of ground deformation monitoring results were obtained,and the accuracy was verified by leveling data.The result showed that the deformation rate obtained by SF method was more accurate,the standard deviation of the difference from the leveling data was 5.58 mm/a,and the deformation accuracy was improved by 35.57%comparing with the temporal-spatial baseline thresholding method,which verified the effectiveness of SF method in interferograms selecting.