Efficient Net: A Deep Learning Framework for Active Fire and Smoke Detection

Abstract

In this paper, we propose a video-based model for fire detection using a model designed to detect fire and smoke after video processing. Then, the model was developed by increasing the rate of fire detection in a single image and using a pre-trained model. The real-time detection procedure is verified in 0.1 second. Also, an AI technique has been created to detect smoke and fire using deep learning (Effective Network). This is a more stable and faster technology than the current technologies in use. Like VGG16, VGG19, ResNet and the comparison was made with ResNet because it is better than other techniques. The results indicated that the proposed technique was better than ResNet.