1) Forest fires are often caused by heat generated in litter that ignites accidentally from human carelessness in summer. Current detection methods rely on air, temperature, and particle sensors that must be triggered by particles entering the sensors. 2) The proposed method uses image processing techniques to detect fires from images captured by a digital camera. It converts RGB color space to YCbCr color space to separate luminance from chrominance and better identify fire pixels. 3) A convolutional neural network algorithm will be used to analyze the images for fire detection. If detected, a warning will be triggered to prevent damage to forests, wildlife, and resources from wildfires.