Development and Evaluation of a UAV-Photogrammetry System for Precise 3D Environmental Modeling.

Area of Science:

• Geomatics Engineering
• Photogrammetry
• Remote Sensing

Background:

• Existing Unmanned Aerial Vehicle (UAV)-photogrammetry systems often lack specialized solutions for development requirements.
• Adequate assessment of UAV-photogrammetry system implementation has been insufficient in recent research.
• Precise modeling of complex environments like open-pit mines requires robust and validated photogrammetric approaches.

Purpose of the Study:

• To present the methodological and experimental implementation of a specifically designed UAV-photogrammetry system.
• To propose solutions for enhancing system performance and the quality of photogrammetric outputs.
• To evaluate the accuracy of the developed system in modeling a challenging open-pit gravel mine environment.

Main Methods:

• System development involved an electric helicopter, high-resolution camera, and inertial navigation system (INS).
• Custom in-house software was developed for camera/platform calibration, system integration, data acquisition, flight planning, and self-calibration.
• Accuracy assessment involved direct and indirect georeferencing with varying ground control points (GCPs), imaging configurations, and network stability.

Main Results:

• Direct georeferencing achieved 1.55 m horizontal and 3.16 m vertical absolute modeling accuracy.
• Indirect georeferencing significantly improved accuracy to 0.4 cm horizontally and 1.7 cm vertically.
• Analysis confirmed the impact of imaging configuration and network stability on overall modeling precision.

Conclusions:

• The developed UAV-photogrammetry system offers effective solutions for precise environmental modeling.
• Indirect georeferencing substantially enhances absolute accuracy compared to direct georeferencing.
• The system's comprehensive approach addresses critical requirements for advanced UAV-based mapping applications.