Drones have learned to determine emergency
Czech engineers have equipped the drones with special systems that recognize potentially dangerous situations. Such devices should help the police.
Researchers have created a system that can analyze video footage from drones. Then it is passed to analysts so that they can determine whether the behavior of the people being filmed is dangerous or not.
Neural networks are used to decode the received data. They help not only to control the behavior of the crowd or individuals, but also the operation of technical means. For example, the movement of cars, the work of traffic lights, and so on.
All footage is divided into parts, analyzed and collected into a single picture. After that, what is happening is compared with those close to ideal conditions. With regard to behavior, there is a comparison of actions in a standard environment and the existing anomalies. If there are any.
One of the advantages of the system is learning. She is also capable of performing actions in real time. This helps the police to make decisions without wasting time.
During the drone tests, the development analyzed the actions of football players on the field. During the match, the players were given a signal to lie down on the ground, and the system immediately signaled an anomaly. All data was recorded and analyzed.
It is assumed that the method of Czech scientists will allow law enforcement agencies to respond more quickly to emergencies.
Other developments, albeit not of the most recent years, can be found on the OXO website in the section "Technics".
Researchers have created a system that can analyze video footage from drones. Then it is passed to analysts so that they can determine whether the behavior of the people being filmed is dangerous or not.
Neural networks are used to decode the received data. They help not only to control the behavior of the crowd or individuals, but also the operation of technical means. For example, the movement of cars, the work of traffic lights, and so on.
All footage is divided into parts, analyzed and collected into a single picture. After that, what is happening is compared with those close to ideal conditions. With regard to behavior, there is a comparison of actions in a standard environment and the existing anomalies. If there are any.
One of the advantages of the system is learning. She is also capable of performing actions in real time. This helps the police to make decisions without wasting time.
During the drone tests, the development analyzed the actions of football players on the field. During the match, the players were given a signal to lie down on the ground, and the system immediately signaled an anomaly. All data was recorded and analyzed.
It is assumed that the method of Czech scientists will allow law enforcement agencies to respond more quickly to emergencies.
Other developments, albeit not of the most recent years, can be found on the OXO website in the section "Technics".
