GCP-Based Automated Fine Alignment Method for Improving the Accuracy of Coordinate Information on UAV Point Cloud...

Area of Science:

• Geomatics Engineering
• Computer Vision
• Robotics

Background:

• 3D point cloud data (PCD) is crucial for construction, often captured by unmanned aerial vehicles (UAVs).
• UAV photogrammetry frequently results in inaccurate PCD, limiting its application.
• Improving the coordinate accuracy of PCD is essential for reliable construction applications.

Purpose of the Study:

• To develop an automated framework for enhancing the coordinate accuracy of UAV-based PCD.
• To integrate image-based deep learning and PCD analysis for precise geometric data.
• To validate the framework's performance and efficiency compared to existing methods.

Main Methods:

• A four-phase framework: Ground Control Point (GCP) detection, GCP global coordinate extraction, transformation matrix estimation, and fine alignment.
• Integration of image-based deep learning and PCD analysis techniques.
• Experimental validation including fine alignment performance and comparison with Iterative Closest Points (ICP).

Main Results:

• The proposed framework achieved millimeter-level accuracy along each axis.
• The system demonstrated a run time of less than 30 seconds.
• The framework significantly improved the coordinate accuracy of UAV-based PCD.

Conclusions:

• The developed framework offers an automated and efficient solution for improving PCD coordinate accuracy.
• The millimeter-level precision and speed indicate high feasibility for construction applications.
• This advancement supports more reliable use of UAV-derived 3D data in the construction industry.