EDGE AI BASED FOREST FIRE DETECTION SYSTEM

Forests are a vital natural resource that directly influences the ecosystem. Recently, forest fires have become a serious issue due to natural and man-made climate effects. To address this, an artificial intelligence-based forest fire detection method is proposed for smart city applications, aiming to prevent major disa1sters through early detection. This research reviews vision-based forest fire localization and classification methods. Specifically, it utilizes a forest fire detection dataset to tackle the classification problem of distinguishing between fire and no-fire images. The research introduces a deep learning method named FFireNet, which leverages the pre-trained convolutional base of the MobileNetV2 model, supplemented with fully connected layers to address the specific task of forest fire recognition. This model classifies images as either depicting a forest fire or not, based on extracted symmetrical features. The performance of FFireNet in classifying fire and no-fire images was evaluated using various performance metrics and compared with other convolutional neural network (CNN) models. The results demonstrate that the proposed approach achieves a classification accuracy of 98.42%, an error rate of 1.58%, a recall of 99.47%, and a precision of 97.42%. These outcomes are promising, highlighting the effectiveness of FFireNet in solving the forest fire classification problem using a unique forest fire detection dataset.