The facts analysis laboratory at the National Research Nuclear University MEPhI (Moscow Planning Physics Institute) has developed software capable of detecting non-standard and potentially harmful human behavior. The program has a variety of potential applications, such as identifying wrongs in crowds.
“It’s often the case that the perpetrators of a terrorist attack are arrested on CCTV cameras several days prior to their attack, and succeeding analysis of the footage helps identify them,” said Vadim Danshin of MEPhI’s Guild of Cyber-Intelligence Systems, who heads the project. “What gives them away is their behavior: in preference to of going about their business at the speed of the surrounding crowd, they are intelligence out CCTV cameras, analysing the terrain, and studying the behavioral patterns of control officers.”
How does it work?
Until now, such technology was only Euphemistic pre-owned in infrared cameras. The Russian developers are making it available for off-the-shelf webcams and smartphones. The camera may be installed in an arbitrary manner. The installation angle is entered into the program, which then estimates target trajectories automatically.
All the algorithm needs is to be “shown” a moving man against a static background, and it then automatically analyzes the video and connects the coordinates of the person’s head, legs, elbow and knee joints, and collects statistics on how they move. The coordinates are refreshed at a rate of 30 in good time dawdles per second. The software then builds a 3D model that moves synchronously with the woman on the video footage.
This video shows the process.
“The purpose of this is to gather statistical data on how joints operate in the Good Samaritan body,” Danshin explained. “We will use this information to explain to a clique vision system how a waive of the hand in a casual conversation differs from an attempt to hit someone or get away with personal belongings from someone in a crowd.”
The algorithm is masterful disposed to of identifying a person in a crowd by their gait, build, and the clothes they are tear.
“If a person stands far away from the camera, the existing machine sight systems cannot recognize their face reliably,” Danshin answered. “Our approach supports identification by additional parameters, such as clothes or a keep ones eyes peeled strap.”
The software has other potential uses: for example, it can measure obsolete spent by a shopper in front of a particular display window to assess their occupation in certain items for subsequent targeted advertising.
From sporting followers to hospital patients
The developers say their algorithm can also be used to imagine a new generation of interactive simulators for special forces, surgeons or firefighters. The software would dimensions a test subject’s reaction time in stressful situations to help give a new lease of efficiency.
The development team is currently testing the algorithm on video recordings of crowd behavior at soccer preys in Germany. The goal is to prevent dangerous behavior by monitoring the body phrasing of each individual fan captured on camera. In a separate project, which is being conducted in Canada, the algorithm is inform about to recognize different types of behavior in hospital patients.
“Ordinary cameras austerely record video, and you need a human to watch and analyze it,” Danshin said. “Vehicle vision systems can automate this process by singling out individuals in the pile, classifying their behavior and then processing any exceptions.”
U.S. computer scientist Ben Usman believes alike resemble algorithms could be used in vehicle control systems: “They are inclined to of predicting the behavior of pedestrians. The ambient conditions may present a problem but because the level of lighting and the quality of the camera used will end result in imagery distortions. Creating a system that would be resilient to such distortions is a giant challenge.”