Intelligent Surveillance System for Real-Time Radioactive Materials Localization and Identification Based on Fusion of Vision and Radioactive Spectroscopy

This research presents an innovative surveillance system capable of real-time detection, identification, and localization of radioactive materials within video feeds from surveillance cameras. By combining advanced image processing algorithms and gamma-ray spectroscopy, this system offers a novel approach to the challenge of radioactive material detection. Image processing techniques are employed to identify and track objects, such as bags or packages, that may contain radioactive materials. Simultaneously, gamma-ray spectrometers measure the energy spectra emitted by radioactive sources, enabling the identification of radionuclides through spectral analysis. Data fusion techniques integrate the information derived from image processing and spectroscopy, yielding accurate and reliable localization and identification of radioactive materials. Experimental results demonstrate that detector placement and distance from the radioactive source significantly influence the system's performance. In a recent experiment, the system identified and localized a contaminated radioactive source containing Cs-137 (100 ci) within 20 seconds and 5 seconds, respectively. This approach offers a cost-effective and efficient solution for radioactive material detection and localization, surpassing conventional methods. With broad applications in world security, nuclear facility monitoring, and environmental surveillance, this innovative system empowers timely and informed decision-making in the face of nuclear threats and incidents, enhancing preparedness and response capabilities.