Several of the successors to Google Glass have auto-focus cameras of reasonable resolution. We have developed novel methods to read text in photographs and would apply this in the Smart Glasses project. The wearer of the glasses would issue a voice command “read” which would result in a synthesized voice reading the text in view. Optical character recognition is a well-established technology for scanned images of documents. It is far more challenging to read documents using a camera, which is not held perpendicular to the plane of the text, where lighting conditions vary and where the image contains other artefacts. However, we have developed effective solutions to these problems and our technology could be applied in this device. As wearable technology advances, offering higher resolution cameras, Internet connectivity and increased memory, it will become possible to read smaller print, for example newspapers and to optionally translate this. It is also envisaged that the device could integrate with other systems, for example allowing a wearer to ask for a bill to be paid, a telephone number to be dialled or a letter generated. We are not aware of a similar product that is currently available and have been advised that our approach incorporates five patentable innovations.
Insight & Impact
The idea applies to all the above areas, although I seem only able to tick one box. Partially sighted people would be able to read labels on foodstuffs, their correspondence and bills, instructions on packaging and products, notices and timetables and so on. This would result in an enormous increase in a person’s capacity to lead a normal life and would greatly reduce the stress resulting from their inability to read. It would also have beneficial health and safety results, as warnings and instructions could be understood.
Skills needed for this project
- We would wish to collaborate with a body such as the RNIB to trial the device and to gain feedback from focus groups.