Intelligent Itinerant Inspection Technology of Aircraft Based on Deep Learning and AR

Itinerant inspection of aircraft is an important link to ensure the smooth flight of aircraft at high altitude. At present, the inspection of civil aviation aircraft mainly depends on human visual inspection, which is greatly affected by the environment, time-consuming and labor-intensive, and has many defects. To solve these problems, this paper proposes an intelligent itinerant inspection method based on deep learning and AR. Collect relevant pictures of aircraft defects and expand them to form a data set. Use the yolov5 network to train and learn the characteristics of various types of aircraft defects, form a specific depth learning model, and integrate it into AR glasses to identify aircraft defects. Use augmented reality to display the identification results. Finally, the Boeing 878 aircraft was taken as the object for experimental verification, and the aircraft defect recognition rate reached 80%, which verified the effectiveness of this aircraft intelligent itinerant technology.