Extraction of text on TV screen using optical character recognition

Abstract

Optical character recognition (OCR) is a very active area of research and has become very successful in pattern recognition. It is based on algorithms for machine vision and artificial intelligence and used in developing algorithms for reading text on images, e.g. reading registration plates, scanned books and documents, etc. This paper presents the system for text extraction on the image taken by grabbing the content of the TV screen. The preparation steps for OCR are developed which detect the text regions in an image. An open-source algorithm for OCR is then run to read the text regions. The system is used as a part of Black Box Testing system in order to test and functionally verify TV set operation. After reading the text regions, comparison with the expected text is performed to make a final pass/fail decision for the test case. The system successfully read the text from the TV screen and can be used in a functional verification system. © 2012 IEEE.