SUAI has developed an intelligent monitoring system that allows for the automatic detection of signs of vandalism in historical parks and gardens.
St. Petersburg’s historical gardens and parks are complex ensembles that include monuments, small architectural forms, decorative fences, and elements of landscape design. These areas are visited by hundreds of people every day, and not all of them treat the city’s heritage with the proper respect. Graffiti on walls, mechanical damage to sculptures, trampling of lawns, and littering are just a few of the problems faced by security services.
«Traditional video surveillance systems record what is happening, but the effectiveness of such control depends directly on the human factor. An operator is physically unable to analyze dozens of video streams simultaneously, and many offenses are short-lived. As a result, a significant number of incidents go unnoticed,» says Nikita Vasilets, a graduate of the SUAI Institute of Information Technology and Programming and the developer of the system.
The solution he proposes is based on computer vision and machine learning technologies and is an intelligent system that can automatically detect signs of potential vandalism.
The key difference between this system and traditional video surveillance systems is its autonomous analysis capabilities. It does not require continuous operator involvement, as the video stream is processed automatically and alerts are generated only when specific conditions are met. This approach has the potential to reduce the workload on employees and shorten the response time to potential incidents.
