These software modules for video processing and recognition are designed for distributed video surveillance systems, encompassing "Smart Home" and "Smart City" infrastructures.
Purpose
- Detection, recognition, and tracking of people, including those moving along complex trajectories
- Analysis of human activity and detection of specific events (e.g., falls, sudden acceleration, movement in prohibited directions, etc.)
- Person re-identification - detecting the same person in different locations and/or at different times in images from distributed CCTV systems
- Early detection of primary fire indicators - smoke and flame
- Monitoring of available parking spaces in car parks
Main advantages:
- High accuracy of video data analysis and recognition
- Automatic adaptation to different surveillance conditions
- Efficient utilization of modern processing methods, including proprietary developments
- Reduced video data processing time through support for parallel computing on graphics processing units (GPUs)
Контакты:
Rikhard Bogush
Euphrosyne Polotskaya State University of Polotsk
29 Blokhin Street, Novopolotsk
211440
Belarus
Tel.: +375 214 42 30 31
