Abstract
Optical character recognition is the process of converting characters from image format to text format. The process includes four main stages namely: pre-processing, feature extraction, character recognition, and post-processing. The success is mainly based on the feature extraction method and the character recognition algorithm. When optical character recognition systems are developed for mobile devices, two main constraints must be addressed. They are the system’s size and speed. In this research, both were considered. A small feature dataset was created using polar histogram of x and y projections of the character image. In addition, a simple character recognition algorithm based on Euclidean distance was adapted. An accuracy of over 99.9% was achieved in near microsecond scale recognizing execution time on the development system.
Highlights
Optical character recognition (OCR) is the process of character recognition where a character in image pixels format is converted to character text format such as ASCII or Unicode [1], [2], [3]
There are a huge number of features that can be used in OCR systems [6]
A dataset with 1016 fonts for each character was used for feature extraction that was based on the polar histogram of the x and y projections of character image
Summary
Optical character recognition (OCR) is the process of character recognition where a character in image pixels format is converted to character text format such as ASCII or Unicode [1], [2], [3]. The process starts by preprocessing the text image that contains characters to prepare it in a form that can be used to extract unique features. Once the features are obtained, the recognition algorithm is chosen and implemented. In many situations, postprocessing such as output formatting may be needed These are the four main OCR system stages [4]. There are three main goals of any OCR system: speed, accuracy, and storage capacity [1]. Because of the storage and processing speed limitation, mobile devices have some constraints on the OCR applications. There are many surveys for general OCR and for language specific OCR [10], [11], [12]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
