Survey on Forest Wildfire Detection Using Deep Learning

Abstract

Abstract—The advent of satellite technology has made it possible to continuously monitor and manage forest fires, which pose a serious hazard to people and other living things. Smoke in the air indicates the presence of forest wildfires. Fire detection is essential in fire alarm systems for preventing damage and other fire catastrophes that have an impact on society. It's crucial to effectively identify fire from visual settings to prevent large-scale fires. An efficient method of a convolutional neural network based Inception-v3 based on transfer learning is developed to increase the accuracy of fire detection. It trains satellite images to classify datasets into fire and non-fire images, generates a confusion matrix to determine the framework's effectiveness, and then uses local binary patterns to extract the fire-occurring region from satellite images. This method lowers the rate of false detection. Keywords–:: Convolutional Neural Network (CNN), deep learning, Inception-v3, fire detection, image classification.

DCP-Net: An Efficient Image Segmentation Model for Forest Wildfires

“…The disk space occupied by FireNet is only 7.45 MB, and it can run steadily at a frame rate of 24 frames per second, achieving over an accuracy of over 93% on experimental datasets. Jitendra Musale et al [26] developed an efficient method based on transfer learning using the convolutional neural network Inception-v3, which divides the dataset into fire and non-fire images by training on satellite images. Zhang et al [27],…”