Smart devices for visually impaired

There is a lot of work and research being done to find ways to improve life for partially-sighted and blind people. Reading and recognition devices could make smartphones, tablets, and smart glasses into indispensable aids for the visually impaired.

CINVESTAV (Center for Research and Advanced Studies of the National Polytechnic Institute) in Mexico, there is another smart glasses project in progress. It combines computational geometry, artificial intelligence, and ultrasound techniques, amongst other things, to create a useful aid for the visually impaired a

A light weight, ergonomically acceptable prototype it almost looks like a normal pair of glasses and can work in real time with batteries that last approximately four hours in continuous use.

Image: CINVESTAV

The prototype combines glasses with stereo sound sensors and GPS technology attached to a tablet, which can give spoken directions and recognize denominations of currency, read signs, identify colors, and other things. It also employs machine learning to recognize different places and objects. Because it uses ultrasound, it can also detect translucent obstacles, like glass doors.

Braille ebook reader

Like a Kindle for the blind.

Image: Fraunhofer Institute http://www.anagraphs.eu/

The Anagraphs project took up the idea and began to work on plans for a device that would employ thermo-hydraulic micro-actuation to activate Braille dots by infrared laser radiation via a micro-mirror scanning system. It’s easier to imagine it as a kind of wax material, which can go from solid to liquid with heat and be easily reshaped to create Braille dots.

FingerReader

The technology for reading written text continues to improve and the FingerReader is a good example of a new way of interacting. This MIT Media Labs project is a wearable device, a very chunky ring that sits on the finger and is capable of detecting and interpreting 12-point printed text as the user scans his or her finger across it. It reads aloud in real-time. Small vibrations alert the wearer to any deviation off the line.

Image: Fluid Interfaces Group MIT Media Lab

It’s currently just a concept with a prototype, but it has potential applications beyond the visually impaired for teaching children to read or translating languages. We already have apps capable of doing this on our smartphones, and OCR (optical character recognition) is getting fairly reliable, but the FingerReader provides a more natural way of interacting.

ARIANNA app

Sometimes the best solutions are the ones that take advantage of the technology you already have. There are a few great apps available for the blind and partially sighted, such as TapTapSee, which can recognize objects that you take a photo of and tell you what they are.

The ARIANNA app (pAth Recognition for Indoor Assisted NavigatioN with Augmented perception) solves another difficult problem. GPS and other navigation systems tend to struggle with indoor environments. The name of the app is a reference to Greek mythology and a clue to how it works. Before it can be used you must stick colored tape to the ground to mark out specific routes, much like you see in hospitals.

Image: arxiv.org/abs/1312.3724 : ARIANNA: pAth Recognition for Indoor Assisted NavigatioN with Augmented perception

Users of the app point their smartphone camera at the ground, and as they wave it back and forth there’s a vibration when it finds the line. This means that there’s no problem with audio interference, and people can use the device without headphones. As an extra layer, it’s possible to place QR codes that give additional information, like telling the user there’s a water fountain nearby, or a toilet.

If you’re worried about the need for colored lines everywhere then you’ll be glad to know the researchers have thought of that. They suggest infrared paths, which can be picked up by smartphone cameras, but remain invisible to naked eye, as a possible way to create paths in future.

For more details visit : Anagraphs project

pAth Recognition for Indoor Assisted NavigatioN with Augmented perception

https://www.media.mit.edu/

 

Comments are closed.