An integrated fire detection system using IoT and image processing technique for smart cities

•The novelty of system is real-time monitoring, early prediction, validation through UAV and fire confirmation using image processing.•The proposed system presents higher true fire detection rate of about 95-98 percent.•This study is first of its kind that integrates UAV for the confirmation of fire event and improves real time monitoring.•Simulation results have shown promising outcomes for accurate fire detection and same may be utilized for smart cities.•The proposed solution may also be extended for addressing the need for many more disaster management activities in smart cities. In the current scenario, the concept of Smart Cities is one of the emerging and challenging research areas. The cities are surrounded by forests, agricultural land, or open areas, where fire incidence can occur threatening human life and causing many resources to become extinct. This article aims to design an early fire detection system to get rid of fire events using the concept of senor network and UAV’s technology. The architecture proposal is based on sensors for monitoring environmental parameters and to process the information through sensors and IoT application. The proposed fire detection system is the combination of wireless sensor technologies, UAVs, and cloud computing. Some image processing techniques are also integrated into the proposed fire detection system to identify the fire event with better accuracy and used as an integrated solution. To improve the true detection rate, rules are also designed. The simulation results of the proposed fire detection system are compared with several existing methods. It is observed that the proposed system has a higher fire detection rate to improve the true detection of forest fire from 95 to 98 percent.