Imagine standing in front of several road signs but unable to locate the one you want, because they are all written in a foreign language. This is the job for the OTON GLASS, a device to capture the image, translate it to a language of your choice, and read it to you in your ear. Not only will this help travelers abroad, but also help people with poor vision and those suffering from dyslexia.
Oton Glass is the effort of Keisuke Shimakage, who says he was inspired to develop the device by his father’s dyslexia. Keisuke got together a team of engineers and designers from the Media Creation Research Department at the Institute of Advanced Media Arts and Sciences, Japan, and started on the project.
At the heart of the Oton Glass is a Raspberry Pi 3 (RBPi3), along with two tiny cameras, and an earphone. One camera resides on the inside of a spectacle frame, tracking the user’s eyes. As soon as it detects the user blinking with no eyeball movement, another camera on the outside of the frame captures the image of whatever the user is looking at. The RBPi3 then processes the image, running it through an optical character recognition program. If there are any written words in the image, the RBPi3 coverts them to speech, and plays it through the earphone into the user’s ear.
Although the initial prototype of the Oton Glass was slow in capturing and replaying the text into audio, the team was able to cut down the time from 15 seconds to a mere 3 seconds in their second prototype.
The team designed the case in CAD software and 3-D printed it to be able to test it in real life situations. With feedback from dyslexic users, they were able to upgrade the device further.
At present, the Oton Glass is doing the rounds at several trade and tech shows throughout Japan, and is ready for public distribution. Trial is underway with models of the device at the Nippon Keihan Library, Kobe Eye Centre, and the Japan Blind Party Association. The Oton Glass has won the runner-up prize for the James Dyson Award of 2016. It has also generated huge attention at several other award shows and in the media.
In front of the inside camera of the Oton Glass is a lens with a half mirror that reflects the eye of the user, which the camera tracks for movements. The outer camera waits for a trigger from the blink of a still eye resulting from the wearer reading something, and captures the image and passes it to the RBPi3.
The RBPi3 uses an optical character recognition software to filter out any characters in the image. It then uses artificial voice technology to change the words into sounds, whose meaning the user can understand. If the Oton Glass is unable to recognize some characters, it sends them to a remote server to decipher. This allows the Oton Glass to translate anything that the user sees. The device combines camera-to-glasses and looks very much like normal glasses.
