An Internet of Things (IoT) Assisted Forest Fire Detection Mechanism Using Hybrid Classification and Learning Model

For all the primary and secondary advantages they provide, forests are an indispensable part of humanity's natural resource arsenal. Forest fires and other natural catastrophes significantly affect both the rate of global warming and the viability of Earth's biota. Consequently, studying how to automatically detect forest fires is crucial for reducing the frequency and severity of such catastrophes. Determinants of fire prevention and suppression strategies can benefit from early detection as well. In this study, we use computer vision algorithms based on artificial intelligence to identify fire and smoke in photographs. This paper presents a new hybrid methodology called Enhanced Learning with Fire Classification Model (ELFCM). It uses learning in conjunction with classification norms to accurately report fire and smoke details. To test how well the proposed scheme works, it is cross-validated with the conventional Random Forest Classifier (RFC). According to the findings of the experiments, some of the unreported or incorrectly detected fires that ELFCM reported may be confirmed by using IoT devices. The fire department receives this recorded report for their records. In the end, we tested the fire classification findings by comparing our method's performance to that of other previously announced fire detection methods that employed commonly used performance matrices.