Scientists from the University of Glasgow are working on a smartphone that can create images from sound after developing a machine learning algorithm to measure echoes and sounds capable of generating images. The technology constructs an image of the device's surroundings, which produces sounds without the need for cameras. The tool can be mounted on any device with a microphone and speakers or radio antennas. Dr. Alex Turbine stated, "What sets this research apart from other systems is that it requires only a microphone or antenna to create 3D images. The algorithm we've developed can turn any device with a microphone or antenna into a sonar locating device."
To create the images, the advanced machine learning algorithm measures the time it takes for sound emitted from speakers or radio waves from small antennas to bounce back inside a space and return to the sensor. By analyzing the results, the algorithm can deduce the shape, size, and design of the room, identify the objects or individuals within it, and then present the results in the form of a video stream that converts echo data into 3D visualization. Researchers say this technology could aid in monitoring buildings without the need for traditional cameras, tracking movements of vulnerable patients in nursing homes, and monitoring a patient's chest movements to detect any changes in their breathing, according to the Daily Mail.
