Valeriy L. Popov - senior lecturer of the department ‘Criminal Law, Criminal Procedure and Criminalistics’ of the Law Institute of the Russian University of Transport
Abstract. Current enforcement of transport security cannot do without the use of artificial intelligence based on neural network technologies. In 2019 the Decree of the President of the Russian Federation approved the National Strategy for the Development of Artificial Intelligence for the period up to 2030, according to which this becomes a priority direction of the state policy of Russia for the next 20 years. For these purposes, in April 2020 in Moscow, there has been established a special legal regime for five years, which allowed the use of personal biometric data of a person without his consent, which made it possible to launch a smart video surveillance system to identify a face in the Moscow Metro, implemented by VisionLabs, which identified each person entering and exiting the metro by his/her appearance, and using the FacePay service one could make and pay for travel only by a face image. At the same time, biometric identification and human authentication, made by artificial intelligence, directly coincide with the tasks of forensics. Based on this, the methods for achieving goals in neural networks and forensics were compared and analyzed from the point of view of scientific validity and reliability of the obtained results. The current paper has analyzed the possibilities of using neural networks in the forensic trials on transport.
Keywords: neural networks; artificial intelligence; forensics/expertise; biometrics; identification; authentication; transport security/safety; metro.