Development of a System for Detecting Production Parts within the Framework of Technological Processes in the Aviation Industry

The aviation industry faces ongoing demands for safety, reliability, and efficiency, driving continuous technological advancements. Ensuring high-quality components is crucial, with a focus on developing systems for detecting and locating aircraft parts within production processes. These systems enhance production efficiency and reduce error risks. Various technologies, including machine vision, laser triangulation, and ultrasound testing, have been explored, but the unique characteristics of aviation technology necessitate specialized approaches. Artificial Intelligence and Machine Learning offer promising solutions, enabling independent image analysis and continuous performance improvement. Integrating these systems into existing production processes automates quality control throughout the entire cycle. This study proposes a system for detecting aircraft components during the painting process using computer vision and neural networks. The system's architecture is designed, tested with real-world data, and optimized for accuracy and speed. A user interface is developed for easy interaction, and the system is integrated into a production process for evaluation. This approach aims to create a highly efficient system for automating quality control, increasing production efficiency, and reducing defects. The paper discusses the current challenges, analyzes existing methods, and presents a detailed architecture for the detection system, highlighting its potential to significantly enhance the quality and dependability of aviation equipment.