.png){kind=icon}
.png){kind=icon}
Return
The influence of photogrammetric point-cloud quality in direct cloud-to-FE modelling and simulation of architectural heritage structures: a case study of a Hindu temple
.png){kind=icon}
PRE
AI
A
I
I
I
DOI：10.26833/ijeg.1754285.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Photogrammetry-derived point clouds are widely used as a geometric foundation for assessing architectural heritage structures. Traditionally, this process is conducted through a multi-step workflow, specifically the scan-to-intermediary-surface-to-finite-element-model (FEM) approach. Recently, a newer method-direct point cloud to FEM (Cloud2FEM)-has enabled a more efficient single-step workflow. This approach requires a point cloud as its primary input. However, the influence of specific characteristics of the point cloud source (e.g., photogrammetry-derived) on the Cloud2FEM process and its performance remains unclear. Therefore, this study aims to investigate this influence and recommend data quality standards to enhance the automation rate of Cloud2FEM-based structural analysis. Close-range drone photogrammetry was employed to document the Apit Temple. The acquired data were processed using a Structure-from-Motion pipeline, and point cloud quality was assessed through root-mean-square error (RMSE) and multiscale model-to-model cloud comparison (M3C2) distance analyses. During the Cloud2FEM conversion, two key evaluation steps were introduced: centroid generation and polyline-to-polygon conversion. The performance of the resulting FE model in dynamic analysis was then evaluated using seismic data from the Athena earthquake. The point cloud quality assessment indicated RMSE values ranging from 1.1 cm to 5 cm and a mean M3C2 distance of 5.53 mm. The two key evaluations revealed a strong association between the quantity, distribution, and completeness of point clouds and (a) the accuracy of planimetric shape representation via centroids, and (b) the success rate of polyline-to-polygon conversion. The developed FE model accurately simulated structural dynamics and identified three vulnerable nodes, consistent with previous reports. These findings emphasize that the quantity, distribution, and completeness of point clouds are critical parameters in achieving accurate and efficient Cloud2FEM modeling. Accordingly, meticulous photogrammetric surveys are essential to generate high-quality point clouds for heritage structure simulations. This highlights the importance of high-quality point clouds for accelerating FEM-based structural modeling and analysis.
Keywords:
Photogrammetry
Point-clouds
Scan-to-FEM
Architectural Heritage
Structural Analysis
Journal
I
IF:
2.5
Papers:
160
Citations:
348
Organization
Cited Papers