The offset-tracking method (OTM) utilizing SAR image intensity can detect large deformations, which makes up for the inability of interferometric synthetic aperture radar (InSAR) technology in large mining deformation monitoring, and has been widely used. Through lots of simulation experiments, it was found that the accuracy of OTM is associated with deformation gradients and image noises in the cross-correlation window (CCW), so CCW sizes should be selected reasonably according to deformation gradients and noise levels. Based on the above conclusions, this paper proposes an adaptive CCW selection method based on deformation gradients and image noises for mining deformation monitoring, and this method considers influences of deformation gradients and image noises on deformations to select adaptive CCWs. In consideration of noise influences on offset-tracking results, smaller CCWs are selected for large deformation gradient areas, and larger CCWs are selected for small deformation gradient areas. For some special areas, special CCWs are selected for offset-tracking. The proposed method is implemented to simulation and real experiments, and the experiment results demonstrate that the proposed method with high reliability and effectiveness can significantly improve the accuracy of OTM in mining deformation monitoring.

To meet the needs of the rapid development of new energy sources, China is currently accelerating the construction of pumped storage power stations (PSPS). However, the complex geological environment and large-scale construction can easily cause deformation areas such as collapse and sliding on the surface of PSPS. This paper adopts ground-based interferometric synthetic aperture radar (GB-InSAR) technology to monitor surface deformation using 206 ground-based radar (GBR) data collected from 2022/07/31 10:24 to 2022/08/06 14:57 with a PSPS in the construction stage in western China. In the selection of permanent scatterers (PSs) for GBR data processing, statistical homogeneous pixels (SHPs) information is considered in this paper. Based on the fast SHP selection (FaSHPS) algorithm, we introduce intensity deviations ( ${D_I}$ ) to exclude pixels with unstable intensity performance in the temporal dimension and propose the ${D_I} + FaSHPS$ algorithm, which is experimentally proven to have higher SHPs selection accuracy and computational efficiency. By combining with the traditional PSs selection algorithm, the distribution density of monitoring points is effectively improved. In addition, based on the deformation results and on-site survey, this paper analyzes and discusses the identified deformation areas in detail from the aspects of natural geography and man-made activities. The results show that the surface is relatively stable overall, except for the effects of man-made activities. This study has demonstrated the remarkable reliability and superiority of the GB-InSAR technology in identifying and analyzing deformation areas in the construction stage of PSPS.

The complex terrain and abundant ravines in the western mountainous areas of Beijing have led to dramatic changes in the geological environment. Monitoring and assessing the stability of landslide surfaces is of great significance for disaster prevention and ensuring the safety of the capital city. According to the spatial similarity characteristics of landslide surfaces, we propose a new interferometric synthetic aperture radar (InSAR) deformation enhancement method by taking into account the time-series deformation information of spatially adjacent homogeneous monitoring points. Taking Dongjiang Gully, Beijing as a typical study area, using multi-temporal InSAR technology, 80 scenes of RADARSAT-2 data from September 2016 to September 2022 were processed to obtain their time-series surface deformation to verify the advantages of the proposed method. The results show that the standard deviation of the deformation difference of all monitoring points is generally reduced after the deformation enhancement, and the mean value is reduced from 5.1 to 3.3, which is 35.2% lower in comparison. Then this study assesses the stability of the landslide surface based on the deformation enhancement results. Firstly, the optical image interpretation was combined with the angular distortions derived from deformation gradients to analyze the spatial location and boundaries of the landslide, and then to identify the infrastructure that is more susceptible to landslide impact. Secondly, through the principal component analysis (PCA) method, the correlation between each component and the distribution characteristics of surface deformation was analyzed. Finally, starting from geological factors and triggering conditions, the driving force for landslide surface deformation was discussed, and it was drawn that seasonal precipitation is a major influencing factor. The proposed method can provide a reference for landslide monitoring and assessment in similar areas.