ResNet-50 based fire and smoke images classification
Hedi Jabnouni, Imen Arfaoui, Mohamed Ali Cherni, Moez Bouchouicha, Mounir Sayadi
Abstract
Fires are becoming a greater threat to people’s lives, property, and the environment. Image technologies, as compared to traditional fire detection approaches, hold a lot of promise for overcoming the problem of a high false alarm rate. However, one of the key drawbacks of these systems is their time-consuming and labor-intensive creation. In fact, multi-feature techniques, such as chromatic characteristics, dynamic features, texture features, and contour features, are frequently used to create implemented algorithms. Therefore, we provide, in this paper, a study of some transfer learning model, and we compare it to a proposed model based on convolution neural network (CNN) algorithm. To do this, we consider a proper database composed by a total of 28334 images classified into three categories: 7329 fire images, 9205 smoke images and 11800 other images.