Abstract
Visual impairment is one of the major limitations to a human. Communication and information access is one of the challenges for this people. Most of information is communicated with the text message. Text is available in the forms of books, newspapers, digital screens, commercial boards, sign boards etc. Now a day, worlds become more digital, with the availability of digital technology. There is a need of digital aid to assist the visual impaired people on the problem of recognizing printed text in the any image such as scene or printed image. In the motive of supporting visual impaired, a system assists such people by reading out the printed or handwritten text to this people. This system is also applicable to normal human beings who read the text as a speech as possible. In this system, we want to design a device that captures an image through camera, extract only part of image that contains only text and converts that text content into speech. This system is constituted by Raspberry pi 3B+, high definition camera, headphone and power supply. The captured image go through a variety of image pre-processing techniques to detect only that region of the image that holds the text and removes the background. To convert the text image to speech two tools are used. They are image pre-processing techniques and TTS (Text-to-Speech) synthesizers. The aural output is heard through the speakers or earphones.
Full Paper