A group of researchers at the University of Glasgow has developed a machine learning algorithm that uses reflected echoes to create 3D images of the environment.
According to the inventors, it can be used, for example, to catch robbers or monitor patients without using cameras.
The system analyzes sounds from speakers or radio waves emanating from antennas. The algorithm measures how long it takes for these signals to bounce off the wall and return to the sensor.
He then calculates the shape, size and layout of the room, and determines the presence of objects or people. The system converts the received data into three-dimensional images in the form of a video stream.
The developers compared the principle of operation of the algorithm with the senses of bats and stated that it can be run on any smartphone or laptop.
“Firstly, creating 3D images requires data from just one input – a microphone or an antenna. Second, we believe that the algorithm we have developed can turn any device with one of these elements into an echolocation device, ”said lead author of the study, Dr. Alex Turpin.
Scientists also believe that their algorithm will significantly reduce the cost of creating 3D images.
Recall that in April 2021, researchers at the Massachusetts Institute of Technology created a robot that can see objects through walls using radio frequency identification.
In March, the FDA approved the first AI device to detect latent signs of COVID-19 in asymptomatic vectors.
In February, Google introduced technology to measure heart rate and respiratory rate using a smartphone camera.
Subscribe to ForkLog news on Telegram: ForkLog AI – all the news from the world of AI!
Found a mistake in the text? Select it and press CTRL + ENTER