Integration of Convolutional Neural Networks into Technical Education Through Automated Visual Quality Inspection in Additive Manufacturing

The development of artificial intelligence and intelligent manufacturing systems creates new opportunities for the automation of technological processes while increasing the demand for digital and technological competencies in engineering education. Particular importance is attributed to the integration of machine vision and deep learning methods, which enables theoretical knowledge to be connected with practical problem-solving in modern manufacturing environments. This study investigates the application of convolutional neural networks to automated visual quality inspection in additive manufacturing, with particular emphasis on their integration into experimentally oriented engineering education. The research aims to develop and experimentally evaluate an approach that combines image data processing, artificial intelligence methods, and automated quality assessment of manufactured components within a unified application environment. Particular attention is given to the potential of neural networks for identifying visual manufacturing defects and to the factors affecting classification reliability under practical operating conditions. The proposed approach provides a framework for investigating the relationships between image acquisition conditions, input data quality, and classification outputs. Furthermore, it supports the practical understanding of machine vision, deep learning, and intelligent manufacturing principles through engagement with real technological data. The main contribution of this study lies in the design and implementation of an experimental platform that systematically integrates additive manufacturing, machine vision, CNN-based classification, and experimental evaluation into a unified framework for practice-oriented engineering education.