The device, called RF Capture, which was founded at MIT, is able to do the impossible – to “see” through walls and recognizes human figures, gestures and movement. “Voyeur” use waves of WiFi, which send a special transmitter, and then measure the reflected signal to the sensors.
In this way, as explained in Tech Insider , created what might be called the image the person who is behind the wall.
You need a supercomputer
Fortunately, some time will pass before anyone (including a neighbor) will be able to buy such a device. Transforming signals to the image requires the enormous computing power. It is necessary to distinguish the signal reflected by the man of the signals reflected by other objects and sources of radiation.
As you can see in the picture below, at first glance it is difficult to conclude that the group spots on the screen represents the man. However, the algorithms can learn to recognize several key features that will give the belief that the tracked object is a person.
Gizmodo says that the tests give a 90% success rate in identifying the man.
The project is still in the experimental stage, however, can soon be used even in the search for victims of natural disasters and accidents.
No comments:
Post a Comment