🧪 Test?View on arXiv
UAV Thermal Imagery for Inert Ordnance Screening: Multi Campaign Dataset Development,Object Detection, and Practical Recommendations
object detectionthermal imagerydataset developmenthumanitarian demining
2609.01738
Builder Relevance
2h ago70%
Abstract
This study developed a multi campaign UAV thermal image dataset for detecting unexploded ordnance (UXO) and provided practical recommendations for humanitarian mine action.
Reality Card
Core Claim
The study created a dataset of 5,855 thermal image label pairs and evaluated object detection models to improve UXO screening processes.
Method / Result
The final dataset included 420 training and 106 validation images from a 33 m flight altitude and 629 training and 157 validation images from a 15 m flight altitude.
Limitations
The study emphasizes the need for qualified human review, which may limit automation and reproducibility.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.