Face Recognition System in St. Petersburg

Infrastructure of the agro-industrial complex "Safe City" of St. Petersburg

• Main article: Information technology in St. Petersburg
• Infrastructure of the agro-industrial complex "Safe City" of St. Petersburg
• Video surveillance system in St. Petersburg

2021: St. Petersburg enveloped a facial recognition video surveillance network. It can expand to 100 thousand cameras

In St. Petersburg, it is planned to significantly expand the use of video identification technology for individuals and create prerequisites for creating an end-to-end urban-scale video identification system. This was told to TAdviser by Acting Chairman of the Committee on Informatization and Communications of St. Petersburg Vladimir Tumarev.

As of February 2021, almost 18 thousand city cameras are used for video identification of persons: they are located at the entrances to the front doors of residential buildings, in places where mass cultural events are held, Tumarev cited data. And in total in the city more than 46.5 thousand cameras.

We plan to gradually increase this figure to 180 thousand by 2025, as well as expand their presence in places of mass crowding of people, at objects of transport and social infrastructure of the city. It is planned to use most of these video cameras within the framework of the video identification system of persons, which will create the prerequisites for an end-to-end urban video identification system, which includes more than 100 thousand city video cameras, "says the acting chairman of the committee on informatization and communications of St. Petersburg.

Cameras with video identification technology in St. Petersburg are located at the entrances to the front doors of residential buildings, in places where mass cultural events are held "(photo - video.tcb-spb.ru)"

An epidemic of a new coronavirus gave impetus to development in this area. According to Vladimir Tumarev, during the period of coronovirus restrictions in the spring of 2020, the committee organized interaction with the Moscow Information Technology Department (DIT) as soon as possible. 17 thousand cameras of the city video surveillance system of St. Petersburg were connected to the video identification system of persons to search for violators of the self-isolation regime who arrived from abroad.

This allowed law enforcement agencies and Rospotrebnadzor to stop a significant number of such cases. According to the results of the test use of the video identification system, positive feedback was received from law enforcement agencies about the system, with the help of which a significant number of wanted persons were identified and crimes were solved, says Tumarev.

The committee on informatization and communications did not specify what facial recognition technology is used in the video surveillance system. The Moscow video analytics system is multivendor, told earlier in DIT: as a result of the competitions, several suppliers were identified, their algorithms complement each other. One of these vendors is the Russian startup Ntechlab.

In September 2020, Ntechlab said that in pilot mode they launched a video surveillance system with facial recognition, which is used in Moscow, in ten more large cities, including Nizhny Novgorod.

Back in 2017, St. Petersburg tested a video surveillance system that can recognize faces using the NtechLab solution (see the block below).

2017: Solution Testing FindFace

Since the beginning of October 2017, testing of a video surveillance system capable of recognizing faces starts in St. Petersburg. The FindFace program, to which all street surveillance cameras will be connected, will be able to analyze the image obtained from them in real time and identify offenders.

To ensure law and order, police officers will use the so-called means of objective control to video record possible offenses. The videos allow you to quickly identify the offenders and bring them to justice, "said Vyacheslav Stepchenko, head of the information and public relations department of the Main Directorate of the Ministry of Internal Affairs for St. Petersburg and the Leningrad Region.

The system will automatically transmit data on suspects to law enforcement agencies. FindFace can also track the route of criminals, preventing them from hiding even in a large crowd, and identify the instigators of riots at mass events.

The system has access to databases of various departments, including law enforcement agencies. In turn, only law enforcement agencies, special services and municipal structures have access to FindFace data. The technology is based on self-learning neural networks. FindFace is able to recognize faces with an accuracy of 70%, but the developers intend to increase it to 100%. The system easily identifies any user who has a page on the VKontakte social network. Everyone can download the mobile application FindFace.

In September 2017, the Moscow City Hall reported that more than 3 thousand video cameras of the city video surveillance network were connected to the face recognition system. At the time of launch, this is one of the largest security systems in the world, where identity identification is used on this scale, noted in the DIT of Moscow.

Notes