A Human Detection Approach for Burning Building Sites Using Deep Learning Techniques

In fire situations, the main concern of first responders and firefighters is to save people trapped in the burning site. In such environments, one factor making the process of looking for people challenging is having high levels of smoke hindering a clear vision for both people and white light cameras. In this paper, we are proposing an approach to detect people trapped in burning sites, in a manner that will ensure the safety of firefighters and accelerate the rescue process of victims. The proposed approach suggests using Infrared (IR) cameras to capture thermal images from the burning site as they are not distorted with smoke. Deep learning technique using CNN would be applied on thermal images to detect victims in fire situations.