Security
With the proliferation of CCTV cameras, computer vision enables the analysis of public and workplaces, airports and industrial facilities so that security personnel can quickly respond to potential threats.
Face recognition
Face recognition has long been an integral part of the security system of a large number of objects for searching persons who are on the list of unwanted or dangerous persons. However, no less interesting is the case of searching for people who are not on the lists, but may pose a theoretical threat, for example:
-
All persons in a certain age range of a certain gender
-
All persons wearing glasses
-
All persons have seen during a certain time
-
All people that were seen in one place
The presence of these tools allows the security service to act much earlier to prevent possible danger.
Abandoned luggage
Luggage left unattended every day adds a lot of trouble to the airport security service. Baggage without an owner is always perceived by the security service as an immediate threat and requires compliance with all security protocols. Neural networks can automatically search for left luggage and inform airport services if they are found.
Virtual fence
A large number of premises and zones to which entry is prohibited may be accessible to outsiders due to the negligence or inattention of the staff. Solving this problem is possible on the basis of the existing video surveillance infrastructure by adding an additional Virtual fences video analytics module.
Virtual fences allow control of restricted areas and create an alarm in case of an emergency situation. Virtual fences can be set up to work only with the people and ignore personnel in uniform or animals for reducing false alarms.
Weapon detection
Automatic search for persons with weapons in their hands will allow the security services to act with maximum speed to counter the commission of a crime.
Vehicles identification
Recognition of car license plates has found wide use in our lives, it is used in automatic speed control systems on roads or accounting for the time spent in a supermarket parking lot. The development of these systems are automatic VIN code reading devices.
Military vehicles identification
Analysis of materials from video cameras and drones to identify military equipment are tasks of military intelligence and forensics in which deep learning can help people. Identify and classify the type of combat vehicle? Count their number? Highlight and recognize license plates? These and many other problems can be solved by computer vision in automatic mode.
Drone detection
In today's world, drones are a huge danger, they can violate the safety of air traffic, serve as instruments of terrorist acts and even be a weapon on their own. Therefore, the detection of drones is the most important task that can be solved by computer vision. Classification of drones, birds, airplanes and helicopters, here is a short list of tasks that artificial intelligence solves in anti-drone systems.
