Optical Character Recognition from Text Image

Abstract

Optical Character Recognition (OCR) is a system that provides a full alphanumeric recognition of printed or handwritten characters by simply scanning the text image. OCR system interprets the printed or handwritten characters image and converts it into corresponding editable text document. The text image is divided into regions by isolating each line, then individual characters with spaces. After character extraction, the texture and topological features like corner points, features of different regions, ratio of character area and convex area of all characters of text image are calculated. Previously features of each uppercase and lowercase letter, digit, and symbols are stored as a template. Based on the texture and topological features, the system recognizes the exact character using feature matching between the extracted character and the template of all characters as a measure of similarity.