Akademik Veri Yönetim Sistemi
Neural Computing and Applications, cilt.37, sa.24, ss.20105-20126, 2025 (SCI-Expanded, Scopus)
Aircraft engine inspection is a key pillar of aviation safety as it helps to maintain adequate performance standards to ensure engine airworthiness. In addition, it is also vital for asset value retention. Borescope inspection is currently the most widely used visual inspection method for aircraft engines. However, borescope inspection is a time-consuming, subjective, and complex process that heavily depends on the experience and attention level of the inspector. Moreover, the cost savings of airlines and the maintenance, repair, and overhaul (MRO) centers expose pressure and workload on inspectors. These factors make an automated system to support damage detection during borescope inspection necessary in order to mitigate potential risks. In this paper, we propose a deep learning-based automated damage detection framework that employs aircraft engine borescope inspection images. Faster R-CNN-based deep learning model with Inception v2 feature extractor is utilized for the present architecture. Due to the limited number of images, data augmentation and other overfitting methods are also employed. The framework supports crack, burn, nick, and dent damage types across all modules of turbofan engines. It is trained and validated with moderate to complex borescope images obtained from the field. The framework achieves 92.64% accuracy for crack, 92.05% for nick or dent, and 81.14% for burn damage classes, with an overall 88.61% average accuracy.