A Standalone OCR System for Mobile Cameraphones

Abstract

In optical character recognition (OCR), visible characters appearing as images (i.e. on paper) are recognized as symbolic characters and stored in a computer's memory or similar device. The purpose of this work is to find whether current mobile cameraphones are able to run OCR software without relying on dedicated hardware or facilities provided by the network. This article describes a simple OCR system that was implemented in Symbian C++ to be run on a stock Nokia 6630 cameraphone. The system is limited to recognizing English capital letters printed in black, against a white background. The opportunities and hardships related to bringing OCR to run on mobile platforms having image capturing capability are also discussed in more general terms