Mobile Optical Character Recognition

Abstract:

Optical Character Recognition (OCR) is a technology whose main purpose is to scan documents in physical format (whether they're handwritten or machine-written) and convert them into fully search- able and accessible digital format contents. The use of Optical Character Recognition systems is often resumed to document scanning and conversion, however, its application goes far beyond when merged with artificial intelligence. Nowadays, the usage of OCR has become so rooted in different application that most people are not even aware of its existence and usage. In this context, this paper provides an overview of this topic, its state-of-the-art and the research already carried out in this field. More- over, it presents the implementation of an OCR mobile application for Android and its natural language processing features. Results show that using Firebase and ML Vision Text Recognition models it is possible to create a mobile app with oine/online capabilities and near real-time text recognition.