Intelligent Image-Based Detection of Surface Anomalies in Aircraft Structures

In the aviation industry, the structural integrity of aircraft is paramount to ensure safety and reliability. The process of regular inspections and maintenance is integral to identifying abnormalities on the surface of the aircraft, which could cause catastrophic failures if left unchecked. Usually, the traditional process of inspections involves the visual assessment of the aircraft surface by a human, who may take a long time, is prone to inaccuracies and requires abundant knowledge. This work presents a discussion on the use of the YOLO image processing technique for the automated detection of abnormalities that exist on the surface of aircraft, including cracks, corrosion, dents, and delamination. The use of image processing for the detection of aircraft surface abnormalities stands to revolutionise the process of maintenance associated with aircraft. A second stage was added to the process, which focuses on including a human in the loop verification, where the inspector will be able to evaluate and validate the results. The technology possesses the potential for ease of scalability for the assessment of numerous aircraft in the fleet. The process leads to beneficial outcomes, including safety, cost reduction, and improved reliability.