Maxima Telecom will install thousands of facial recognition cameras in Moscow for 1 billion rubles
Customers: Center for Traffic Management of the Moscow Government (GKU Data Center)
Contractors: Maxima Telecom Product: Integrated video surveillance projectsProject date: 2021/07
Project's budget: 1 billion руб.
|
- How the city video surveillance system is arranged in Moscow
- How the facial recognition system is arranged in Moscow
July 23, 2021 it became known about the tender for the installation of 2 thousand CCTV cameras on the roads of Moscow. The competition won the "Maxima Telecom," which will implement the project for 1 billion rubles. The contract was concluded with the Center for Traffic Management (DPC) of Moscow.
According to Kommersant, citing technical documentation for public procurement, the contractor will have to install cameras in three stages until December 2023. One of the requirements for the equipment is that it must continuously transmit a video stream to the customer's servers in high resolution 4K, and it will also be necessary to provide facial recognition, automatic object tracking and area of interest.
The source of the publication said that the new cameras will allow deploying a neural network in the capital, which will be a security system. At the same time, operators will first train artificial intelligence. For example, the robot will record the incident, show the operator, and he will confirm the presence or absence of the accident.
This will be a neural network that will be able to automatically record incidents in the field of view of cameras and notify the relevant services. In the future, the neural network will be able to work completely independently, without the help of operators, a source in the city hall told the publication. According to him, in the future, all roads in Moscow will be covered with such cameras, which could cost the budget about 4-5 billion rubles.
According to experts interviewed by the publication, by July 2021 Moscow has many cameras, but the problem is that operators physically do not have time to view a huge stream of video. A neural network that recognizes actions, including incidents, will allow you to get more useful information from video in addition to algorithms that are already used, said NtechLabs founder Artem Kukharenko.[1]