TECHNICAL APPLICATION OF IMAGE PROCESSING IN THE IDENTIFICATION OF FOREIGN OBJECTS IN THE AEROSPACE INDUSTRY

In the aviation industry, foreign objects (FOs) - any items not part of the aerospace product - pose risks of damage if they come into contact with aircraft. This study explores the implementation of both automated and manual image processing techniques for the detection of FOs, aiming to enhance safety by reducing inspection times and improving quality over traditional manual methods. The proposed approach utilizes infrared cameras for image capture in low-light or complex environments, combining automated detection algorithms with manual oversight. Experiments were conducted on the structures and cockpit of the Super Tucano 314, Xavante AT-26, and the F5 (J85 model) engine, employing the Yoosee application for imaging and Python with Anaconda Spider for processing. The results demonstrate the system's ability to identify FOs that are difficult to detect visually, ultimately contributing to a reduction in operational risks within the aerospace sector.