Mr Daha Albakari, a final-year Physics student at the Federal University Dutse (FUD), has developed an Artificial Intelligence-powered fire detection system.
Albakari said this during an exhibition of the system on Thursday in Dutse.
He explained that the features and applications of the system, known as AI Fire System, combine computer vision, embedded AI, Internet of Things (IoT) and environmental sensing to provide fast and intelligent fire detection.
Albakari said that unlike conventional fire alarms that are activated after detecting smoke or heat, the system continuously analyses visual and environmental data in real time to identify potential fire hazards early.
He noted that conventional fire detection systems could generate false alarms from cooking smoke, steam and dust, and provide limited information for assessing the level of risk.
“The system uses an ESP32-CAM to capture images in real time, while environmental sensors monitor temperature and gas concentration,” he said.
According to Albakari, an embedded AI model analyses the inputs to classify conditions as fire, smoke or normal, and determine the level of risk.
“Once a potential hazard is detected, the system activates local alarms and can transmit notifications through Wi-Fi or GSM connectivity,” he said.
Albakari said that the invention was in response to the need for an affordable and locally developed solution to provide early warnings before minor fire incidents escalate.
He added that the system was designed for use in homes, schools, hospitals, offices, markets, hotels, warehouses, factories, filling stations and other high-risk environments.
Albakari highlighted that the use of affordable hardware and locally developed AI technology could make intelligent fire safety solutions more accessible in Nigeria.
