AI-based Human Detection and Localization in Heavy Smoke using Radar and IR Camera
- Resource Type
- Conference
- Authors
- Kulhandjian, Hovannes; Davis, Alexander; Leong, Lancelot; Bendot, Michael; Kulhandjian, Michel
- Source
- 2023 IEEE Radar Conference (RadarConf23) Radar Conference (RadarConf23), 2023 IEEE. :1-6 May, 2023
- Subject
- Communication, Networking and Broadcast Technologies
Computing and Processing
Fields, Waves and Electromagnetics
Robotics and Control Systems
Signal Processing and Analysis
Location awareness
Training
Animals
Radar detection
Radar
Cameras
Real-time systems
Deep learning
human detection in heavy smoke
data fusion
artificial intelligence
- Language
One of the main challenges currently firefighters are facing in search and rescue operations is battling the heavy smoke inside a space that needs to be searched for people and animals. In this work, we develop an integrated system composed of two unique sensing mechanisms that are capable of real-time detection and localization of humans and animals in deep smoke to improve the situational awareness of firefighters on the scene. We make use of data from a micro-Doppler sensor and an infrared camera and train a DCNN algorithm to localize a human in dense smoke in real-time. Experimental results reveal that the proposed system can detect a human in heavy smoke with an averaae of 98 % validation accuracy.