Cloud based Text extraction using Google Cloud Vison for Visually Impaired applications

As number of visually impaired people increased year by year due to accidents and biological disorders. Hence there is a need for assistive device that can helpful for their day to day activities. Hence, with recent advancement in the technology, we intend to implement assistive device for visually impaired person such smart reader that is capable of capturing an image from a camera and extract the text from the captured image and further to convert the text to speech as voice based output to assist the visually impaired people. The captured image is analyzed using Google Cloud Vision API Optical Character recognition (OCR). In order to extract text, we use image preprocessing methods to remove any noise or blur in the captured image so that the accuracy can be increased. Further, we include software-based text to speech to convert the text to speech as voice output. In this project we use Raspberry pi interfaced with Night vision camera and speaker to produce the converted text as voice. By applying Google Cloud vision OCR with preprocessing results in high accuracy level of the extracted text through image enhancement methods.