Image to Multilingual Text Conversion for Literacy Education
Muhammad Ajmal, Farooq Ahmad, Martinez-Enriquez A.M., Mudasser Naseer, Muhammad Aslam, Mohsin Ashraf · 2018
Currently both language and images are essential to the texts we use while language and images can combine for literacy education. In this research work, an application is developed to translate the text integrated to images for visual literacy. Further, several approaches for image-to-text multilingual translator are reviewed in detail. By overcoming the gaps, which are identified by thorough review of the literature, an improved methodology is proposed. As a result, the development of application goes through four major phases including: capturing, extraction, recognition and translation. Moreover, Optical Character Recognition algorithm is particularly used for character extraction and recognition with high accuracy under different environmental circumstances. It translates text just by capturing an image with user smart phone camera and translation instantly appears on user's mobile screen in language selected by the user. The proposed solution may particularly be helpful in literacy education, for learning different languages and possibly can work as visitors' assistant.