Parallel Processing in OCR - A Multithreaded approach
G.L. Jayavardhana Rama, A. G. Ramakrishnan, Devaraj Gupta · 2002
Document Image processing has been a frontline research area in the field of human-machine interface for the last few decades. Optical character recognition (OCR) of Indianlanguage scripts has been a topic of interest for quite some time. The earlier contributionswere reported in [1] and [2]. More recent works are reported in [3] and [4]. The need forefficient and robust algorithms and systems for recognition is being felt in India, especially inthe postal department for sorting mail and for preserving out-of-print old books by digitisingthem. Character recognition can also form a part in applications like intelligent scanningmachines, text to speech converters, and automatic language-to-language translators. Veryfew OCRs are available commercially for Indian languages. Most of these systems are font-specific and require the knowledge of the font