The extraction of architectural structural line features can simplify the 3D spatial representation of built environments, reduce the storage and processing burden of large-scale point clouds, and provide essential geometric primitives for downstream modeling tasks. However, existing 3D line extraction methods suffer from incomplete and fragmented contours, with missing or misaligned intersections. To overcome these limitations, this study proposes a patch-level framework for 3D line extraction and structural optimization from building point clouds. The proposed method first partitions point clouds into planar patches and establishes local image planes for each patch, enabling a structured 2D representation of unstructured 3D data. Then, graph-cut segmentation is proposed to extract compact boundary contours, which are vectorized into closed lines and back-projected into 3D space to form the initial line segments. To improve geometric consistency, regularized geometric constraints, including adjacency, collinearity, and orthogonality constraints, are further designed to merge homogeneous segments, refine topology, and strengthen structural outlines. Finally, we evaluated the approach on three indoor building environments and four outdoor scenes, and experimental results show that it reduces noise and redundancy while significantly improving the completeness, closure, and alignment of 3D line features in various complex architectural structures.

Automated recognition of rock mass discontinuities in vegetated high-slope terrains remains a challenging task critical to geohazard assessment and slope stability analysis. This study presents an integrated framework combining close-range UAV photogrammetry with an Improved Superpoint Transformer (ISPT) for semantic segmentation and structural characterization. High-resolution UAV imagery was processed using an SfM–MVS photogrammetric workflow to generate dense point clouds, followed by a three-stage filtering workflow comprising cloth simulation filtering, volumetric density analysis, and VDVI-based vegetation discrimination. Feature augmentation using volumetric density and the Visible-Band Difference Vegetation Index (VDVI), together with connected-component segmentation, enhanced robustness under vegetation occlusion. Validation on four vegetated slopes in Buyun Mountain, China, achieved an overall classification accuracy of 89.5%, exceeding CANUPO (78.2%) and the baseline SPT (85.8%), with a 25-fold improvement in computational efficiency. In total, 4918 structural planes were extracted, and their orientations, dip angles, and trace lengths were automatically derived. The proposed ISPT-based framework provides an efficient and reliable approach for high-precision geotechnical characterization in complex, vegetation-covered rock mass environments.

The deformation monitoring of the ring rockfill dam in pumped storage power stations is of great significance. Traditional monitoring techniques such as geodetic survey, GPS, and multi-point displacement meters have high precision and reliability but are limited by point monitoring in terms of layout density and range. InSAR technology has advantages like high precision, large range, all-weather, non-contact, and low cost, yet faces challenges from factors like spatio-temporal decorrelation, atmospheric delay, and vegetation cover. This research utilized the permanent scatterer InSAR processing technology with multi-reference point baseline network adjustment and high-precision DEM data to monitor the surface deformation of the ring rockfill dam in the upper reservoir of Zhanghewan Pumped Storage Power Station. It analyzed the impact of DEM resolution on PSInSAR monitoring accuracy and verified the accuracy of InSAR deformation monitoring using ground synchronous monitoring data from a high-precision measuring robot. The results indicate that the dam and slope of Zhanghewan Power Station's upper reservoir showed an overall uplift trend during the observation period, which was preliminarily judged to be caused by the temperature rise from winter to summer. The correlation coefficient between the monitoring point deformation rate obtained by the InSAR technology and the ground synchronous observation result was 0.838, with an RMSE of ±7.24mm/yr. The higher the precision of the external DEM, the higher the InSAR monitoring accuracy, with an improvement range of 2 - 3mm. By combining the ground and satellite monitoring results with the water level and temperature observation data, it was found that for the ring rockfill dam, the cumulative displacement of the monitoring points was significantly correlated with the temperature, but the displacement change was not significantly correlated with the temperature change. The influence of temperature on the displacement of monitoring points was slow and nonlinear, and different monitoring points had different responses. The cumulative displacement of monitoring points had a weak correlation with the water level, while the displacement change had a stronger correlation with the water level change. The water level had a greater impact on the upstream and downstream displacement of specific points. This study provides an important reference for the research and application of InSAR deformation monitoring of large-area structures such as ring rockfill dams.

The extraction of architectural structural line features can simplify the 3D spatial representation of monitored objects, reduce the storage burden of massive point clouds, and provide crucial geometric parameters for subsequent 3D modelling. To address the problem of line segment discontinuity caused by point cloud quality, this paper proposes a line segment extraction algorithm based on planar graph-cut segmentation. Specifically, a region growing and merging algorithm is employed to segment the point cloud into multiple planar patches. Subsequently, 3D-2D projection is performed to generate grayscale images, from which line segment parameters are identified and extracted using a graph-cut segmentation algorithm combined with vectorization techniques. Then, depending on spatial 3D-2D back-projection, all extracted 3D line segments are optimized by the structural regularization to further mitigates the discontinuity among collinear line segments and provide high-quality linear features for subsequent 3D reconstruction.