According to traditional concepts, the movement of overlying strata and surface damage caused by coal mining in horizontal coal seams are symmetrical in terms of spatial distribution. However, in a lot of engineering practices, this symmetry has not been discovered. We often use the symmetry function to establish the profile prediction function of the surface damage, which results in a large difference between the prediction result and the actual situation. To solve this problem, this paper takes subsidence velocity as an example. Firstly, the spatial distribution functions of subsidence velocity on both sides were deduced theoretically. Through comparison, it is found that the change rate of the spatial distribution curve of the coal pillar side subsidence velocity is smoother than that of the goaf side and the subsidence velocity curves are skewed to the left. Secondly, based on the idea of lossless propagation of harmonic waves and idealizing the propagation environment, the spatial propagation relationship of surface subsidence velocity in the time domain is established. Then, the Box–Cox transform function is introduced to improve the normal distribution probability density function, and a new dynamic subsidence prediction model based on the Box–Cox transformation is obtained, which is suitable for the full mining stage. The model is tested by practical cases, the prediction accuracy is better than 7%, and the prediction results can meet the needs of engineering prediction accuracy (10%). The results of this research can enrich the existing subsidence prediction theory and provide theoretical and technical support for the prediction of dynamic surface damage caused by similar mining.

Most experimental studies use unimodal data for processing, the RGB image point cloud cannot separate the shrub and tree layers according to the visible vegetation index, and the airborne laser point cloud is difficult to distinguish between the ground and grass ranges, to address the above problems, a multi-band information image fusing the LiDAR point cloud and the RGB image point cloud is constructed. In this study, data collected from UAV platforms, including RGB image point clouds and laser point clouds, were used to construct a fine canopy height model (using laser point cloud data) and high-definition digital orthophotos (using image point cloud data), and the orthophotos were fused with a canopy height model (CHM) by selecting the Difference Enhancement Vegetation Index (DEVI) and Normalised Green-Blue Discrepancy Index (NGBDI) after comparing the accuracy of different indices. Morphological reconstruction of CHM + DEVI/NGBDI fusion image, remove unreasonable values; construct training samples, using classification regression tree algorithm, segmentation of the range of the burned areas and adaptive extraction of vegetation as trees, shrubs and grasslands, tree areas as foreground markers using the local maximum algorithm to detect the tree apexes, the non-tree areas are assigned to be the background markers, and the Watershed Transform is performed to obtain the segmentation contour; the original laser point cloud is divided into chunks according to the segmented single-tree contour, and the highest point is traversed to search for the highest point, and corrected for the height of the single-tree elevations one by one. Accuracy analysis of the vegetation information extracted by the method with the measured data showed that the improved method increased the overall recall rate by 4.1%, the overall precision rate by 3.7%, the overall accuracy F1 score by 3.9%, and the tree height accuracy by 8.8%, 1.4%, 1.7%, 6.4%, 1.8%, and 0.3%, respectively, in the six sampling plots. The effectiveness of the improved method is verified, while the higher the degree of vegetation mixing in the region the better the extraction effect of the improved algorithm.

In order to reveal the secular deformation evolution and strain accumulation progress along subduction zone, we focus on high-precision Global Positioning System (GPS) data processing, GPS time series analysis and regional spatiotemporal filtering. The GPS position time series is modeled as a combination of the long-term plate movements, short-term noise and the frequency-dependent variations and tectonic deformation information. Common Mode Errors (CME) is removed and the deformation signals is then extracted from the residual time series by Principal Component Analysis (PCA). Combing the filtered displacement array and the dynamic regional strain array derived from GPS baseline, a spatiotemporal tectonic evolution map is exhibited obviously. We process IGS time series over the 2002-2011 period in Japan island and reveal the dynamic evolution of crustal deformation along subduction zone. The results show that the regional reference frame transformation by constraining one site location and one stable GPS baseline direction can further improve the Signal-to-Noise Ratio (SNR) of GPS observations. A integrated analysis combining the baseline length with azimuth change can show the crustal motion feature more comprehensively. The observed behaviors agree well with the simulation experiment results of the rock rupture in the laboratory. We divide the pre-seismic deformation into four stages, the stable linear strain accumulation stage, the formation stage of the local locked area, local decoupling stage, and strain release stage. We also give an appropriate dynamical model for the interplate coupling to explain the observed deformation characteristics. Research result provides some of the observational new evidence for inter-seismic deformation anomaly detection and the medium- to longer-term seismic hazard assessments.