Commercial city centres with millions of inhabitants and visitors and as many cars, face particular challenges when it comes to policing to increase safety and security and maintain peace on the streets. After all, you can't put a police officer on every street corner. Or can you?
Increasingly, CCTV camera systems are taking the place of the patrolling officer, and becoming their eyes, and even ears, on the street. But whereas in the past, there may have been a controller on every camera, these days it is increasingly important to put intelligence into these complex systems through software, so that the camera networks can support and alert the teams responsible for security when needed, without requiring constant human surveillance.
Nowhere is this approach more apparent than in the city of Sofia, the capital of Bulgaria in the Balkans. Sofia is home to around 2 million people and a million cars. In the past 10 years the population has doubled and along with it, the number of vehicles on the streets, dramatically increasing the load on the city's police force. As a result, in 2005, the Ministry of Interior contracted to implement a fully automated CCTV system to support the policing of the city.
The goal was to cover all major city intersections, all entrance/exit roads to the city, the ring road around the city, the centre and the major public places with an intelligent real-time operating system. The system integrates CCTV cameras, number-plate recognition, face recognition, speed limit and red-light violations registration, centralised traffic lights management and more.
The initial project has covered an area of around 2 km in the heart of the city with more than 100 CCTV cameras, 60 number-plate recognition cameras and six face recognition cameras. The system monitors seven major road intersections and most of the main government buildings.
The solution uses Verint's NextivaTM platform for all live video monitoring, recording and analytics. Proprietary software integrates the Nextiva video solution with the systems supporting number-plate recognition and face recognition and performs other intelligent functions including red-light violation, stolen car alerts and traffic light management. There are five operator stations in the data centre. Each operator is responsible for a specific region, covered by about 20 cameras and including one or two intersections.
The Verint Nextiva-based intelligent functions and video analytics mean the operators can stay focused on the important events, since they are alerted by the system when action is needed. For example, whatever else they are observing at the time, the system can alert the operator when a stolen car passes through their zone, allowing the operator to monitor the situation, verify the vehicle identity, identify possible routes the vehicle may be taking and pass vital data to the police ‘on the ground', all whilst maintaining their view of the situation they were originally observing.
The system also handles off-line processing of traffic infringements, recording and passing relevant information and still photographs to traffic police for further processing.
Feedback indicates the Sofia pilot project is very successful. Police records show that the reported crimes in the area of the pilot project have decreased by 70% since the project commenced. Drivers receiving traffic tickets in the mail tend to pay extra attention while passing through the monitored zones and are more careful not to commit infractions. Information from the system has also proved very useful in cases of collisions, non-traffic violations, crimes, emergency situations and public events.
As a result of this success, the project in Sofia is set to develop beyond the initial phase and is planned to expand over the next two to three years to cover about 200 road intersections, major routes through the city centre and some public places.
