Portable Camera Based Assistive Text and Product Label Reading from Hand Held Object by Using Android App for Blind Person

Abstract

We propose a camera-based mechanical man app. This app helps the blind persons to browse the text on explicit objects. In this system the camera captures the actual text on the object. Multiple techniques square measure applied to its text. Such as Optical Character Recognition that supply the operation of scanning and recognition of text and a few have integrated voice output. From a grayscale image, thresholding are often accustomed produce binary pictures i.e. image with solely black or white colors, Filtering are often accustomed cut back the noise of image, Next image segmentation technique is employed to perform the method of partitioning a digital image into multiple segments. The goal of segmentation is to modify and/or amendment the illustration of a picture into one thing that’s a lot of significant and easier to analyses. Image scaling is the method of resizing a digital image. Next technique employed in this project is template matching. Temples matching is a way in the digital image process for locating tiny components of a picture that match a template image. Also template extraction are often employed in producing as a vicinity of internal control, some way to navigate a mobile golem or as some way to notice edges in images then finally voice output are going to be generated then blind man will simply listen to the text on it explicit object.