Generally, the blind peoples have several visual complications in doing their daily actions. So, the computer vision structure is used to developing the visually weakened life eminence. Unfortunately, the world is not contains any type of previous intermediate or interface. The proposed system is able to retrieve the text contents from the images and processes them with various edges. The anticipated system is used to engender the speech as several text documents. This document results with an effectual processing of text and segmenting of text from the images to support in actual text detection. The anticipated process is executed in the working platform of MATLAB and the consequences are examined by the obtainable processes. © 2018, Springer Science+Business Media, LLC, part of Springer Nature.