摘要：According to the World Health Organization there are approximately 285 million blind people around the world. These people are faced challenges when reading a book. This paper aims to design and implementation new real time mobile cloud audio reading system for blind persons. The proposed methodology consists of following steps: In Client Side, firstly, capture image of text by camera. Secondly, check the page localization. Thirdly, send image to server. In Cloud Side (Server Side), firstly, apply the modify EAST algorithm on received image to text detection. Secondly, apply OCR algorithm to extract text from image. Thirdly, apply post-processing step to correct the in corrected text. Finally, return text to client side to speak it using text-to-speech algorithm.
