Aero-engines damages diagnosis approach based Deep Learning

Abstract

Traditional aircraft engine inspection methods—relying primarily on visual examination and conventional diagnostic tools—are often time-consuming and may lack the sensitivity required to detect all potential forms of damage, thereby posing significant safety risks. To address these limitations, artificial intelligence (AI) techniques, particularly machine learning and deep learning in the domain of

computer vision, have emerged as promising alternatives for enhanc- ing the precision and efficiency of image-based diagnostics. This

study proposes the development of a YOLO (You Only Look Once) deep learning model for the automated detection of aircraft engine damage. The objective is to expedite the inspection process, reduce maintenance costs, and improve the accuracy of damage detection, thereby contributing to safer and more efficient aircraft operations.