Coal mining leads to surface subsidence, landslides, soil erosion and other problems that seriously threaten the life and property safety of residents in mining areas, and it is urgent to obtain mining subsidence information using high-frequency, high-precision and large-scale monitoring methods. Therefore, this paper mainly studies the deformation monitoring of the Datong mining area using Lutan-1 monostatic and bistatic SAR data. Firstly, the latest Lutan-1 bistatic data are used to reconstruct the DSM, and the interferometric calibration method is used to improve the accuracy of the DSM. Then, the surface deformation monitoring of the mining area is implemented by using DInSAR, SBAS-InSAR and Stacking-InSAR with the reconstructed DSM data and Lutan-1 monostatic SAR data. Finally, the deformation monitoring results are compared with the surface deformation results based on the TanDEM data, and both the results are evaluated using the filed leveling data. Taking 20 images covering the Datong mining area as the data sources, the surface deformation results obtained using different InSAR methods in the mining area were quantitatively evaluated and analyzed. The results indicated that: (1) the DSM obtained using the Lutan-1 bistatic SAR data was assessed and demonstrated with the ICESat laser altimetry data an error of 2.8 m, which meets the Chinese 1:50,000 scale DEM cartographic accuracy standard, and the difference analysis with the TanDEM data shows that the terrain changes are mainly distributed in mountainous areas; (2) Due to the improvement in resolution, the registration accuracy of the SAR images and LT-DSM is higher than that of the TanDEM data in the range direction and azimuth direction; (3) Via evaluation with the filed leveling data, it is found that the surface deformation measurement results based on LT-DSM are less affected by terrain, and the accuracy of LT-DSM-SBAS and LT-DSM-DInSAR is improved by 11.5% and 16.3%, respectively, compared with TanDEM-SBAS and TanDEM-DInSAR, which demonstrates the effectiveness of the Lutan-1 bistatic and monostatic data for mine deformation monitoring.

The quality of the interferometric phase of synthetic aperture radar (SAR) is directly related to the calculation accuracy of the digital elevation model (DEM), surface deformation, and other information. However, the traditional phase filtering method must comprehensively consider the characteristics of the interferometric phase in the spatial and transform domains. As it is difficult to consider the diversity and complexity of the interferometric phase, existing methods have problems such as weak denoising and incomplete edge information retention. Therefore, this paper proposes an interferometric SAR (InSAR) phase denoising network based on a multi-attention mechanism (MAPDNet). Introducing channel and spatial attention mechanisms allows the network to use a multi-objective loss function to guide training while considering the edge feature retention of the interferometric phase based on noise suppression. This paper analyzes the performance of MAPDNet on synthetic data sets and real InSAR data. The experiments show that MAPDNet is superior to existing methods in terms of noise suppression and detail retention.

Statistically homogeneous pixel (SHP) selection is an important process in the distributed scatterer interferometric synthetic aperture radar (DS-InSAR) approach. However, prevalent methods struggle to appropriately balance the efficiency and accuracy of selection. To solve this problem, the authors of this study improved the Hypothesis Test of Confidence Interval (HTCI) to propose an adaptive method to select the level of saliency and confidence interval for the HTCI, called Adp-HTCI. The proposed method can accurately select homogeneous pixels while inheriting the high efficiency of the HTCI. Once homogeneous pixels have been chosen, the eigenvalue decomposition of the covariance matrix is used to optimize their phase and perform temporal processing. We used the proposed method along with data on 31 scenes from the Sentinel-1 satellite from 2 June 2021 to 28 May 2022 to monitor the deformation of the surface of the fire zone in the Sikeshu coalfield. The selection results of homogeneous pixels indicate that the proposed Adp-HTCI algorithm can increase the number of selected homogeneous pixels while ensuring the accuracy of the selection results, thereby enhancing the estimation accuracy and reliability of subsequent parameter solving. The DS-InSAR results showed that the cumulative maximum subsidence in the study area within a year reached—138 mm and the point density used by the DS-InSAR approach was 17.28 times higher than that used by the StaMPS approach. The results of cross-analysis with the results of StaMPS verified the accuracy of the DS-InSAR-based approach.

Safe operation of tailings reservoirs is essential to protect downstream life and property, but current monitoring methods are inadequate in scale and refinement, and most reservoirs are built in low coherence areas far from cities. Use of polarization data to monitor deformation may improve area coherence and thus point selection density. With the example of the Kafang tailings reservoir and dual-polarization Sentinel-1 data from 9 August 2020 to 24 May 2021, homogeneous points of different polarization channels were identified with the hypothesis test of the confidence interval method. Results were fused, and BEST, sub-optimum scattering mechanism (SOM), and equal scattering mechanism (ESM) methods were used to optimize phase quality of persistent scatterer (PS) and distributed scatterer (DS) pixels and obtain more detailed deformation information on the area with time series processing. The fusion of homogeneous point sets obtained from different polarization intensity data increased the number of homogeneous points, which was 3.86% and 8.45% higher than that of VH and VV polarization images, respectively. The three polarization optimization methods improved point selection density. Compared with the VV polarization image, the high coherence point density increased by 1.83 (BEST), 3.66 (SOM), and 5.76 (ESM) times, whereas it increased by 1.17 (BEST), 1.84 (SOM), and 2.04 (ESM) times in the tailings reservoir. The consistency and reliability of different methods were good. By comparing the monitoring results of the three methods using polarization data, the hypothesis test of the confidence interval (HTCI) algorithm, and the polarization optimization method will effectively increase the point selection number of the study area, and the ESM method can show the deformation of tailings area more comprehensively. Monitoring indicated deformation of the tailings reservoir tended to diffuse outward from the area with the largest deformation and was relatively stable.