Abstract

In the past, many Braille documents were produced manually on paper, and were reproduced for readers by using methods involving vacuum forming of plastic sheets. However, for some years there has been an increasing trend to use computers for entering, editing and printing Braille documents using special purpose software and printers. This paper describes some work which has been undertaken to optically read the Braille information from old paper copies, and particularly those which have become ‘tired’ and difficult to copy, into the computer so that they can be electronically stored and reproduced. Of specific interest are the techniques of using twin shadows (stereoumbral), i.e. subtracting two pictures of the Braille under different illumination conditions, to overcome the problems of background texture by segmenting this unwanted noise from the important relief information. Initial results for single sided Braille have been found to be promising, and the extension of this approach into the recognition of double sided Braille is discussed.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call