Browsing by Author "Droznika, Mark"

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Aircraft skin machine learning-based defect detection and size estimation in visual inspections
    (MDPI , 2024-09-10) Plastropoulos, Angelos; Bardis, Kostas; Yazigi, George; Avdelidis, Nicolas P.; Droznika, Mark
    Aircraft maintenance is a complex process that requires a highly trained, qualified, and experienced team. The most frequent task in this process is the visual inspection of the airframe structure and engine for surface and sub-surface cracks, impact damage, corrosion, and other irregularities. Automated defect detection is a valuable tool for maintenance engineers to ensure safety and condition monitoring. The proposed approach is to process the captured feedback using various deep learning architectures to achieve the highest performance defect detections. Additionally, an algorithm is proposed to estimate the size of the detected defect. The team collaborated with TUI’s Airline Maintenance Team at Luton Airport, allowing us to fly a drone inside the hangar and use handheld cameras to collect representative data from their aircraft fleet. After a comprehensive dataset was constructed, multiple deep-learning architectures were developed and evaluated. The models were optimized for detecting various aircraft skin defects, with a focus on the challenging task of dent detection. The size estimation approach was evaluated in both controlled laboratory conditions and real-world hangar environments, providing insights into practical implementation challenges.
  • Thumbnail Image
    ItemOpen Access
    Defects recognition algorithm development from visual UAV inspections
    (MDPI, 2022-06-21) Avdelidis, Nicolas Peter; Tsourdos, Antonios; Lafiosca, Pasquale; Plaster, Richard; Plaster, Anna; Droznika, Mark
    Aircraft maintenance plays a key role in the safety of air transport. One of its most significant procedures is the visual inspection of the aircraft skin for defects. This is mainly carried out manually and involves a high skilled human walking around the aircraft. It is very time consuming, costly, stressful and the outcome heavily depends on the skills of the inspector. In this paper, we propose a two-step process for automating the defect recognition and classification from visual images. The visual inspection can be carried out with the use of an unmanned aerial vehicle (UAV) carrying an image sensor to fully automate the procedure and eliminate any human error. With our proposed method in the first step, we perform the crucial part of recognizing the defect. If a defect is found, the image is fed to an ensemble of classifiers for identifying the type. The classifiers are a combination of different pretrained convolution neural network (CNN) models, which we retrained to fit our problem. For achieving our goal, we created our own dataset with defect images captured from aircrafts during inspection in TUI’s maintenance hangar. The images were preprocessed and used to train different pretrained CNNs with the use of transfer learning. We performed an initial training of 40 different CNN architectures to choose the ones that best fitted our dataset. Then, we chose the best four for fine tuning and further testing. For the first step of defect recognition, the DenseNet201 CNN architecture performed better, with an overall accuracy of 81.82%. For the second step for the defect classification, an ensemble of different CNN models was used. The results show that even with a very small dataset, we can reach an accuracy of around 82% in the defect recognition and even 100% for the classification of the categories of missing or damaged exterior paint and primer and dents.