- Статьи
- Internet and technology
- Following digital footsteps: how neural networks help to search for missing people in Russia
Following digital footsteps: how neural networks help to search for missing people in Russia
Russian artificial intelligence (AI) analytics has helped six families in the Novosibirsk Region find their lost children. The Novosibirsk Ministry of Digital Development and Communications and the Ministry of Internal Affairs of Russia for the region were the first in Russia to introduce an AI service for searching for missing children in their region, and this solution has already shown its effectiveness. For more information about how neural networks help to search for missing people and what the prospects of such technologies are, see the Izvestia article
How AI helped to find missing children in the Novosibirsk region
The fact that Russian AI—based analytics helped to find missing children in the Novosibirsk region was reported to Izvestia by NtechLab, a technology partner of Rostec State Corporation. We are talking about the company's development of the AI service Find Face Multi, which analyzes images from city surveillance cameras, focusing on a photo of a missing person.
If his face is found in the human stream, the system informs law enforcement agencies about it. The recognition accuracy is over 99%. Thanks to this, with the help of "smart" video analytics, it was possible to quickly help six families find lost children in the Novosibirsk region.
— The search for a child using biometrics is carried out exclusively with the consent of the parents, — said Anton Loshakov, Deputy Minister of Digital Development and Communications of the Novosibirsk region. — To start the search process, you need to file a missing child report at 112 and then upload his photo through a special digital service.
According to Anton Loshakov, the photo is sent to the secure contour of the Safe City complex for an operational search. As Alexey Palamarchuk, CEO of NtechLab, noted in an interview with Izvestia, due to the flexibility of the product and its ease of use, it can be used as an independent solution or integrated into existing services.
How neural networks help in the search for missing people in different countries of the world
Today, artificial intelligence is actively used to search for missing people both in Russia and abroad, says Alexander Samsonov, a leading expert at the Security Code development and testing department, in an interview with Izvestia. According to the expert, a number of companies in Russia use similar technologies, integrating facial recognition into video surveillance systems.
"There are also successful examples in the world: Clearview AI scans social networks to help law enforcement agencies, Google Person Finder helps reunite people after disasters, and a large—scale facial recognition system is operating in China," says the specialist.
Major AI-assisted search projects for missing people are already underway in the United States, China, and Europe, adds Stanislav Yezhov, Director of AI at Astra Group. In particular, they are used for migration control and criminology purposes. During the Covid-19 pandemic, technology appeared in China that made it possible to identify people who violated quarantine.
These technologies can also be used to find both wanted individuals and missing people. If the algorithms work with the video stream, this will allow them to be identified very quickly, notes Stanislav Yezhov. In turn, Andrei Razdymakho, an expert on cybersecurity at Angara Security, recalls that the Amber Alert system operates in the United States, which integrates AI tools for analyzing video streams and social networks, but the use of facial recognition is limited for ethical and legal reasons.
"In EU countries, the use of such systems is strictly regulated by the GDPR law, and mass face recognition in public places is often prohibited," says the expert.
In Russia, the LizaAlert volunteer search and rescue team uses AI to analyze material from unmanned aircraft during search and rescue operations, said Evgeny Gerasimov, head of the unmanned aviation department of the LizaAlert search and rescue team. According to Izvestia's interlocutor, in 2024, artificial intelligence found 12 people.
What are the prospects for using AI to find missing people?
Human detection technologies based on video and photo information are already very accurate and accessible, Maxim Buzinov, head of the Solar Group's R&D Laboratory, said in an interview with Izvestia. At the same time, in the future, according to the expert, such technologies will only be more accurate and faster: in particular, additional information will appear for search, such as the nature of gait, certain facial expressions or clothes.
— It will be possible to determine, for example, with whom a person communicated or the nature of this communication, — says the specialist. — All this will make it possible to find out the motives and possible intentions of a person.
In turn, Evgeny Gerasimov mentions an increase in the quality of detection, a decrease in the speed of analysis, as well as the launch of data analysis on board unmanned aircraft among the possible prospects for the development of AI search technologies.
Interaction of AI systems with emergency services will be possible not only in connection with missing persons, but also to prevent incidents, adds Tatiana Butorina, an expert consultant and specialist at the Gazinformservice Cybersecurity Analytical Center. It is expected to expand the functionality by processing not only images, but also texts, analyzing social networks and bookmarks to improve the quality of results.
"The integration of AI systems with robots and drones will make it easier to conduct rescue operations to find people even in hard—to-reach and dangerous areas, minimizing risks for rescuers," says Tatiana Butorina.
At the same time, the use of such technologies raises disputes about the balance between search efficiency and privacy protection, recalls Alexander Samsonov. And, although AI has already proven its usefulness in saving lives, society has yet to develop ethical standards for its use.
What problems arise when using AI to find people?
AI-based technologies can indeed help in the rapid search for missing people and many other areas of life, but this creates certain threats related to privacy and privacy secrets, Alexander Lebedev, senior developer of Innostage artificial intelligence systems, says in an interview with Izvestia.
— Now a person will always be in sight and under supervision, whether he wants it or not — this is the price for security, — says the specialist. — As a result, the "Eye of God" system, which connects to any camera and can determine a person's location anywhere in the world, familiar to us from movies and science fiction books, has almost become a reality.
At the same time, the introduction of AI to search for people creates problems not only with privacy, notes Alexander Lebedev. For example, it is not entirely clear what to do when biometric data is compromised. If the data of a passport or other document is in the hands of intruders, you can simply change the document and move on. But if your biometric data is suspicious of the bank or the state, then you will not be able to change it.
On the other hand, organizational difficulties may arise on the way to implementing AI search technologies.: According to Stanislav Yezhov, these include, in particular, integration with existing systems and training of personnel to work with them. The financial side of the issue is also important, since the cost of such systems is quite high. Technologies, of course, confirm their effectiveness, but they require quite serious investments at the moment.
"Legislative issues and interagency cooperation are also a challenge that needs to be addressed both in Russia and in the world," concludes Stanislav Yezhov. — However, despite all the difficulties, technologies still need to be developed, because they help to search for missing persons and identify violators with high enough accuracy and speed.
Переведено сервисом «Яндекс Переводчик»
