The presence of fire is one of the main prerequisites for the occurrence of a fire. Early warning of the occurrence of even a small to larger fire is a good prevention to reduce fire casualties. The paper aims to present a prototype module for real-time fire detection and recognition. The prototype uses concepts of artificial intelligence, neural networks, and computer vision. The use of these technologies is not only useful, but also shows the extent of their development and how useful they are in solving problems of life around us. The module works with augmented reality smart glasses. The modular prototype quickly and in real time detects a fire and sends a message to the user, alerting him of the situation that has occurred. Detecting, recognizing, and signaling when a fire occurs makes a cyber-physical system applicable to the work of firefighters by helping them locate fires from a distance and reducing the likelihood of accidental casualties. The research is based on the study of a lot of data related to the recognition and analysis of fire. Its composition and the reasons for its occurrence are also the subject of research. All accumulated knowledge leads to the successful creation of the prototype.

fire, recognition, computer vision, artificial intelligence, smart glasses, cyber-physical system