AR Guides Implementation for Industrial Production and Manufacturing

摘要

This paper introduces an original approach for the implementation of Augmented Reality automated systems for industrial production and manufacturing. Based on analysis of the up-to-date problems of AR devices practical use like limited usability of AR goggles, low performance and quality of image recognition in real time there is proposed a new approach of accented visualization. The paper addresses some problems of machine vision application for production quality control, which helps solving the challenges for AR guides and improving the efficiency of their implementation in practice. In particular, augmented reality allows users to undergo interactive training and receive active assistance during the execution of technological operations without the need to refer to a set of documentation in paper or electronic form. Within the developed framework there is presented a new solution architecture that combines the capabilities of neural networks, Ontologies and Big Data processing facilities. The proposed approach is recommended to improve the quality of recognizing production operations and manufactured products using machine vision technologies and neural networks in industrial applications.