Multimodal human detection in disaster situations using deep learning & artificial intelligence

Abstract

Mankind has faced natural calamities for survival since the very beginning of human civilization. Even after 65 million years, mankind is still figuring out ways to face natural calamities and survive its post-consequences effectively. Against natural phenomena like- hurricanes, tornadoes, earthquakes, building collapse, forest fires, etc. Humankind is weak and helpless. And no matter how technologically advanced humankind becomes, nature will always remain the strongest opponent that humans have to face for their survival. The revolution of science and technology has helped humankind to invent ways and techniques to survive by fighting against the natural calamities that they face. Technology can reach into places where humans cannot and technology can look deep into details that humans can never go through due to born limitations. Our paper represents the idea of a human detection system that during any calamity, with the help of multiple detection sensors and thermal visual ization techniques, can detect trapped human beings. This human detection system combines the knowledge of Machine learning and Artificial intelligence system tech niques. We hope to contribute to saving human lives during natural calamities and help them to overcome its aftermath in the quickest possible time.