改进小基线集技术的GB-InSAR铁路边坡监测

针对传统小基线集技术形变模型和高相干点识别存在的问题,该文利用基于三重指数串行的目标识别技术选取高相干点,同时提出一种不同于传统奇异值分解的时间序列分析算法。利用振幅离差指数、相干系数均值与强度值三重阈值串行的方案进行高相干点选取;通过阈值合理设定,不仅提高了监测结果的点位分布,也剔除了误选的点,保证了所选高相干点的质量;采用一种符合铁路边坡形变规律的线性模型作为约束条件,解决了时序分析过程中法方程秩亏的问题,并利用最小二乘反演得到了研究区域时间序列形变。实验表明,此方法提高了形变结果的精度和可靠性,证明了SBAS-InSAR技术在铁路边坡形变监测上具有良好的发展前景。

GB-InSAR railway slope monitoring of improved small baseline subset technology

Aimng at the problem of the deformation model and high coherence point identification for traditional small baseline subset techniques,the target recognition technology based on triple exponential serial was used to select high coherent point,and a new algorithm for time series analysis that differs from traditional SVD decomposition was presented.Amplitude deviation index,mean of the coherence coefficients and intensity threshold was used to select high coherent points;Setting a reasonable threshold not only improved the monitoring results of the point distribution,but also eliminated the mistake choose.In order to ensure the quality of high coherent points,a linear model which obeyed the law of railway slope deformation was adopted as the constraint condition to deal with the rank defect of the law equation,the time series deformation of the study area was obtained by least squares.Experiment showed that this method improved the accuracy and reliability of the deformation results,which proved that the SBAS-InSAR technology had a good development prospect in railway slope deformation monitoring.