Forest Fire Detection Using a Low Complex Convolutional Neural Network

Forest fires have significant environmental and societal consequences, contributing to carbon emissions and endangering wildlife. Early detection of forest fires is crucial to reduce the extent of damage and improve firefighting efficiency. This research proposes a computer vision-based approach using a low complex Convolutional Neural Networks (CNNs) for forest fire detection. The CNN architecture consists of three convolutional layers and a fully connected layer with only two neurons, offering a simple yet effective design. The proposed CNN achieves a good accuracy of 97.11% on the Deep Fire dataset, surpassing traditional machine learning and deep learning-based techniques in terms of both accuracy and computational efficiency. The low complexity of the model enables fast processing time, making it suitable for practical real-world applications.