UAV photogrammetry and lidar integration for high-fidelity 3D campus mapping at KFUPM

Area of Science:

• Geomatics Engineering
• Digital Twins
• Smart Cities

Background:

• Smart educational environments require accurate 3D campus models for navigation and facility management.
• Existing 3D mapping methods often lack the detail or operational actionability needed for comprehensive campus management.

Purpose of the Study:

• To develop and demonstrate an end-to-end, replicable Unmanned Aerial Vehicle (UAV) workflow for high-fidelity 3D campus mapping.
• To integrate RGB photogrammetry and LiDAR point clouds for enhanced 3D model construction.
• To evaluate the impact of combined nadir/oblique imagery and super-resolution texturing on model quality.

Main Methods:

• Utilized a DJI Matrice 300 RTK with Zenmuse P1 camera and L2 LiDAR payload for data acquisition.
• Employed nadir grid flights and oblique orbits at 60m altitude with high overlap (80%/70%).
• Co-registered georeferenced LiDAR scans with photogrammetric reconstructions and applied a U-Net super-resolution module for texture enhancement.

Main Results:

• Combined nadir and oblique views improved facade completeness and reduced surface deviation by ~30% compared to nadir-only.
• Super-resolved textures increased Structural Similarity Index (SSIM) from 0.88 to 0.93 and edge sharpness by ~28%.
• The resulting 3D model was successfully exported to a WebGIS for interactive exploration and operational integration.

Conclusions:

• The presented UAV workflow enables the creation of accurate, photorealistic, and actionable 3D campus models.
• Integration of multi-view imagery and LiDAR data, coupled with texture super-resolution, significantly enhances 3D model quality.
• The methodology is replicable and valuable for developing smart educational environments and improving campus management.