Light-Enhanced Deep Learning Approach for Aircraft Visual Inspection Using Cooperative UAVs

Aircraft inspection is a critical aspect of ensuring aviation safety. However, traditional inspection methods are time-consuming, labor-intensive, and lack accuracy, especially under poor lighting conditions. In this paper, we propose a novel light-enhanced deep learning approach for automating airc...

Full description

Saved in:
Bibliographic Details
Published in2025 International Conference for Artificial Intelligence, Applications, Innovation and Ethics (AI2E) pp. 1 - 6
Main Authors Al-Shukaili, Mallak, Al-Abri, Atheer, Al-Busaidi, Renad, Al-Qassabi, Ethar, Al-Gharibi, Yasir, Al-Abri, Said
Format Conference Proceeding
LanguageEnglish
Published IEEE 03.02.2025
Subjects
Accuracy
Aircraft
Deep learning
Drones
feature-based algorithms
Inspection
Lighting
Safety
Surface cracks
Technological innovation
UAVs
visual inspection
Visualization
Online AccessGet full text
DOI10.1109/AI2E64943.2025.10982970

Cover

More Information
Summary:Aircraft inspection is a critical aspect of ensuring aviation safety. However, traditional inspection methods are time-consuming, labor-intensive, and lack accuracy, especially under poor lighting conditions. In this paper, we propose a novel light-enhanced deep learning approach for automating aircraft visual inspection using two drones. One drone (camera drone) captures images and footage perpendicular to the inspection surface, while the other drone (lighting drone) provides illumination from a variable inclined angle to enhance visual inspection and crack detection. We developed advanced deep learning techniques to detect surface anomalies on aircraft effectively. Experimental results on a large-scale dataset demonstrate improved performance compared to existing methods, achieving a high accuracy rate. The proposed method has the potential to significantly improve inspection efficiency, reduce costs, save time, and ultimately enhance aviation safety.
DOI:10.1109/AI2E64943.2025.10982970