Image Text to Speech Conversion using Optical Character Recognition Technique in Raspberry PI

Abstract

Optical Character Recognition (OCR) is a subset of artificial intelligence and is a subset of computer vision. Optical Character Recognition (OCR) is the use of Raspberry Pi to convert scanned bitmap images of handwritten or written text into audio performance. OCRs designed for a variety of world languages are now in use. In this method the context subtraction method based on the Gaussian mixture is used to recover the area of the moving object. For text content, the function of text localization and recognition is used. The text localization algorithm and the Tesract algorithm and edge pixel distributions based on the gradient properties of the stroke directions were used to automatically translate text areas from the object in the Ada enhancement model. In the translated text areas text characters are converted to binaries, which OCR software understands. For the blind, known text symbols are strongly pronounced. The potential of the algorithm for the proposed text location. The text file describes the character codes using the Raspberry system, which recognises the characters by using Tesract's and Python, and the audio output is heard in the recognition step.