81 / 2023-04-04 10:05:57
An Adaptive Coherency Matrix Decomposition Based Polarimetric Persistent Scatterer Interferometry Algorithm for Dual-polarization Sentinel-1 data
Ground Deformation Monitoring,InSAR,Persistent Scatterer Interferometry,Polarimetric Optimization
摘要待审
Leixin Zhang / China University of Mining and Technology;Key Laboratory of Land Environment and Disaster Monitoring, Ministry of Natural Resources,China university of mining and technology
Feng Zhao / China University of Mining and Technology;Key Laboratory of Land Environment and Disaster Monitoring, Ministry of Natural Resources,China university of mining and technology
Sentinel-1 data have been widely employed for monitoring large-scale ground deformation through persistent scatterer interferometry (PSI). However, most studies have only used the VV polarization channel data, failing to take advantage of dual-polarization Sentinel-1 data. The development of polarimetric PSI (PolPSI) methods has made it possible to combine both VV and VH channels to create a new channel, thereby increasing the efficiency and accuracy of the ground deformation monitoring. Traditional high-efficiency PolPSI methods are unable to adaptively optimize persistent scatterers (PSs) and distributed scatterers (DSs) pixels, while currently proposed adaptive methods suffer from high computational burden. To this end, by using dual-polarization Sentinel-1 data, an efficient adaptive coherency matrix decomposition PolPSI (ADCMD-PolPSI) algorithm is proposed in this study. The algorithm firstly separates PS and DS pixels based on the selection results of homogeneous pixels, then employs different phase optimization strategies for different types of pixels. Finally, by combining the optimized PSs and DSs the optimized interferograms are obtained for the ground deformation retrieve. Results in Southern California demonstrate that the proposed ADCMD-PolPSI can effectively improve the interferometric phase quality of the pixels, and, thus, increase the density of high-quality monitoring pixels. Compared to the results obtained using only VV channel data, ADCMD-PolPSI achieves a 494% improvement in the number of high-quality pixels. Additionally, compared to GPS results, the average root-mean-squared error (RMSE) values of ADCMD-PolPSI, the minimum mean square error (MMSE) method (a DS-InSAR method) and VV method are 5.24 mm, 5.33 mm and 5.41 mm respectively. ADCMD-PolPSI presents slightly better performance of ground deformation monitoring accuracy compared to the other two methods.
重要日期
  • 会议日期

    10月26日

    2023

    10月29日

    2023

  • 10月15日 2023

    摘要截稿日期

  • 10月15日 2023

    初稿截稿日期

  • 11月13日 2023

    注册截止日期

主办单位
国际矿山测量协会
中国煤炭学会
中国测绘学会
承办单位
中国矿业大学
中国煤炭科工集团有限公司
移动端
在手机上打开
小程序
打开微信小程序
客服
扫码或点此咨询