Our proposed project mainly focus the object’s text regions. By using artificial neural network, the project offers a novel text localization algorithm. It is prepared by learning gradient features of stroke orientations and distributions of edge pixels. Using off-the-shelf optical character identification software, the text font in the localized text regions is binarized. The renowned text codes are converted into audio output for the blind users. Object distance is measured using ultrasonic sensor. It automatically focuses the text regions from the object. Text extraction is done using neural network based OCR. Text to voice conversion is done using phonematic concatenation for visually impaired people.
Volume 12 | 07-Special Issue
Pages: 593-596
DOI: 10.5373/JARDCS/V12SP7/20202145