.png){kind=icon}
.png){kind=icon}
Return
Efficient and fine-grained viewshed analysis in a three-dimensional urban complex environment
.png){kind=icon}
PRE
AI
DOI：10.1016/j.envsoft.2025.106359.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Performing efficient and fine-grained viewshed analysis in 3D complex urban models, particularly when handling large-scale datasets, presents a significant challenge in Geographic Information Systems (GIS). Existing methods are primarily designed for 2.5D raster models and struggle to effectively manage large-scale data. Furthermore, the commonly utilized approaches for 3D models need large display memory and lack statistical analyses. To address these challenges, this paper adopted a results-oriented approach that diverged from the traditional data-driven paradigm by reformulating the conventional viewshed computation problem as a spatial query problem. Building on this premise, we proposed the Q-View method for oblique photogrammetry data, which involved creating an indexing model for large-scale datasets and enabled parallel querying between the line-of-sight (LOS) and the model. The Q-View method enables efficient and spatially exhaustive analysis, effectively overcoming the complexities associated with traditional viewshed computations. Experimental results showed that our approach achieved a query rate of up to 4 million visibility queries per second on a dataset with 17.6 million triangular meshes. Compared to the latest methods, it offered a 72.45% improvement in operational efficiency and superior accuracy relative to the GPU-rated method. These findings indicated that the proposed method substantially improved both the accuracy and efficiency of viewshed analysis in complex urban environments, providing decision support for urban planning and environmental monitoring.
Keywords:
Spatial viewshed analysis
Spatial visibility analysis
Urban oblique 3D model
Complex environment
Urban planning
Journal
E
IF:
4.6
Papers:
511
Citations:
1.8W