2023 Author: Bryan Walter | [email protected]. Last modified: 2023-05-21 22:24
Japanese and Swiss engineers have created a wearable device that detects finger movements using a compact ultrasonic emitter and receiver glued to the wrist. The device analyzes changes in the acoustic properties of ultrasound caused by the movement of the hand and determines the gesture from them. In addition to gestures with one hand, the system is also capable of recognizing the touch of the fingers of the other hand, say the authors of the article presented at the MobileHCI 2019 conference.
Much of today's consumer electronics is controlled by touch screens, buttons, and other mechanical controls. However, they are not always convenient. Smart watches are a prime example. They have a small screen on which large interface elements cannot be displayed, and the number of physical buttons on them is also usually limited to two or three.
As a solution to the problem, engineers suggest using other methods of entering information. For example, American and Chinese developers have created prototypes of a smartwatch with a projector that turns a hand into a touchscreen. And Google has been working on a mini-radar Soli for several years now, capable of recognizing hand gestures and designed for smartwatches as well as smartphones.
Engineers led by Otmar Hilliges of the Swiss Higher Technical School of Zurich have created a system that recognizes hand gestures using ultrasound. The principle of the system is based on the fact that when performing gestures, the shape of the hand changes. At the same time, along with the shape, the resonance frequencies of the hand also change, so gestures can be recognized by tracking these changes.
The developers have created two separate prototypes of the device (plug-in and stand-alone), but both have a similar design of the main body. It consists of two small, three square centimeters in area, elements fixed to the outside of the brush with double-sided adhesive tape. This pair of piezoelectric elements acts as a speaker, emitting ultrasonic pulses with a frequency of 20 to 40 kilohertz, and a microphone that detects hand vibration.
Wearable device prototype
For analysis, the algorithm turns the data from the microphone into a vector. The vector is then mapped to a particular gesture using a support vector machine. The developers have trained the classifier to recognize many different types of gestures. For example, he is able to recognize the movement of the thumb along the others or to one of them, touching the fingers, as well as touching the fingers of one hand to the palm of the other, and he is able to roughly calculate the location and pressure.
Experiments have shown that gesture recognition accuracy is 84.4 percent. The authors note that while their method has drawbacks, for example, the need to train an algorithm to work with a specific person.
Last year, another group of engineers created a wearable ultrasound gesture recognition device consisting of a bracelet and a ring. The ring with the radiator is located on the thumb, and the bracelet with microphones is located on the wrist. The principle of operation of the system is based on the fact that when passing through the brush, acoustic impulses change their properties.