Thanks to machine learning using AI, automatic monitoring is becoming increasingly effective. The project of British and Indian scientists demonstrated new possibilities of this technology – in particular, it allows to identify aggressive persons with the help of unmanned vehicles in the crowd.
In the article entitled “Eye in Sky”, researchers describe how they with the help of a conventional Parrot AR quadrocopter at a cost of about $ 200 carried video over the mobile Internet for real-time analysis. The algorithm, obtained as a result of intensive machine learning, made it possible to identify aggressively-minded individuals according to the characteristic elements of their poses. Five poses related to violence were chosen for training: strangulation, striking, kicking, shooting and using stabbing objects.
Scientists hope that such algorithms will find their application in solving crimes in public places and during mass events. As an example, the leading researcher at Cambridge University, Amarjot Singh, led the terrorist attack on the Manchester arena in 2017. According to him, such attacks can be prevented if CCTV detect suspicious behavior – for example, leaving the bag for a long time unattended.js.src = “&version=v2.8”; 'script', 'facebook-jssdk'));