ROURKELA: Innovators at the National Institute of Technology-Rourkela (NIT-R) have secured the patent for a fully automated next-generation smart surveillance technology to address the challenge of monitoring unauthorized access in large and complex building environments.
Addressing the limitations of conventional CCTV camera-based surveillance, the new system is fully automated, non-intrusive and can detect, identify and track individuals using integrated thermal imaging technology.
The patented technology titled “Unauthorized person detection using thermal imaging and gait recognition for indoor security” has been developed by Prof Samit Ari of the Electronics and Communication Engineering department along with research scholar Mohammad Iman Junaid and MTech graduates Narayan Prasad Sharma and Irshad Ali.
Members of the research team said conventional surveillance systems based on CCTV cameras require extensive manual monitoring and analysis, making them prone to human error. In large setups, tracking individuals across multiple cameras is both difficult and cumbersome, especially under varying lighting and occlusion conditions. But the new technology is smart, efficient and works on its own, requiring minimal human intervention.
Built-in thermal imaging technology helps distinguish people from the background due to lower infrared noise and enables accurate detection even in low-light conditions. The research team used human gait as a biometric identifier. When a person enters a restricted area, the system compares his/her walking pattern with those of authorized individuals. If no match is found, it marks the individual as suspicious and signals security.
The research team has developed a working prototype with three thermal cameras connected to a central server via USB interfaces. The system detects unauthorized individuals at entry, tracks their movement through multiple checkpoints, maintains a temporary database of unknown individuals, and determines entry and exit patterns based on direction of movement.
