Super Resolution Infrared Thermal Imaging Using Pansharpening Algorithms: Quantitative Assessment and Application to...

Area of Science:

• Remote Sensing
• Image Processing
• Geospatial Technology

Background:

• High-resolution thermal images are limited, hindering fusion with higher-resolution sensors.
• Pansharpening algorithms, successful in satellite multispectral and panchromatic fusion, are explored for thermal imagery.
• Existing pansharpening methods have not been systematically evaluated for unmanned aerial vehicle (UAV) thermal data.

Purpose of the Study:

• To analyze the potential of existing pansharpening algorithms for fusing UAV-acquired thermal images with higher-resolution data.
• To quantitatively compare different pansharpening methods applied to UAV thermal imagery.
• To identify the optimal pansharpening algorithm for UAV thermal image fusion based on objective results.

Main Methods:

• Adaptation and application of established satellite-based pansharpening algorithms to UAV thermal and high-resolution image datasets.
• Quantitative evaluation procedure to compare the performance of various pansharpening techniques.
• Comparative analysis focusing on the fusion of high-resolution imagery with thermal imagery from UAVs.

Main Results:

• A quantitative comparison of pansharpening algorithms applied to UAV thermal images was performed.
• The study provides an objective basis for selecting the most effective pansharpening method for this specific data type.
• Demonstration of successful fusion for heritage documentation applications.

Conclusions:

• The systematic analysis enables objective selection of pansharpening methods for UAV thermal image fusion.
• This research addresses a gap in evaluating pansharpening techniques for UAV thermal data.
• The findings have broad applicability beyond heritage documentation, including various remote sensing fields.