Forest Fire Prediction

Forest fires are becoming more frequent due to human activity and dry weather, making early detection essential to minimize environmental and social damage. While CNN-based fire detection systems offer high accuracy, their heavy processing and memory requirements create challenges for real-world implementation. To address this, we propose a deep learning-based classification model that automates fire detection in images and videos, improving surveillance in industries, hospitals, and other critical areas. This system is designed to balance accuracy and efficiency by analyzing fire data and adapting to different conditions. Additionally, an IoT-powered prediction system using Raspberry Pi and anemometers monitors wind speed to estimate fire spread, enabling quicker response and more effective disaster management.