Assessment of DSM Based on Radiometric Transformation of UAV Data.

Area of Science:

• Geomatics Engineering
• Photogrammetry
• Remote Sensing

Background:

• Unmanned Aerial Vehicles (UAVs) offer high spatial resolution for 3D Earth modeling.
• Assessing the accuracy of low-cost UAV data is crucial for reliable Digital Surface Model (DSM) generation.
• Radiometric transformation techniques can potentially enhance the quality of UAV-derived geospatial products.

Purpose of the Study:

• To evaluate the effectiveness of radiometric transformation on low-cost UAV data.
• To investigate the impact of these techniques on the global and local accuracy of Digital Surface Models (DSMs).
• To compare UAV image-derived DSM accuracy with UAV Light Detection and Ranging (LIDAR) data.

Main Methods:

• Utilized UAV LIDAR data for Ground Control Points (GCPs) and accuracy assessment.
• Processed UAV drone images from 300m and 500m flying heights using Pix4D Mapper.
• Applied radiometric transformation to RAW UAV images acquired at 500m in MATLAB.

Main Results:

• Achieved significant enhancement in DSM accuracy, reducing the Root Mean Square Error (RMSE) in Z from ±2.15m to 0.11m.
• Analyzed DSM accuracy based on point cloud number, density, and distribution.
• Demonstrated compatibility of the enhancement technique with UAV LIDAR accuracy across various land cover types.

Conclusions:

• Radiometric transformation is an effective technique for improving the accuracy of low-cost UAV-derived DSMs.
• The enhanced DSMs exhibit high global and local accuracy, comparable to UAV LIDAR.
• This research validates the use of enhanced UAV image data for precise 3D Earth modeling applications.