Performance Improvements to the BBN Byblos OCR System

Authors:
Michael Decerbo;Premkumar Natarajan;Rohit Prasad;Ehry MacRostie
Affiliations:
BBN Technologies, MA, USA;BBN Technologies, MA, USA;BBN Technologies, MA, USA;BBN Technologies, MA, USA
Venue:
ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
Year:
2005

Citing 4
Cited 2

Heteroscedastic discriminant analysis and reduced rank HMMs for improved speech recognition

Speech Communication
Advances in the BBN BYBLOS OCR System

ICDAR '99 Proceedings of the Fifth International Conference on Document Analysis and Recognition
Videotext OCR Using Hidden Markov Models

ICDAR '01 Proceedings of the Sixth International Conference on Document Analysis and Recognition
Script-Independent, HMM-Based Text Line Finding for OCR

ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 4

Offline Arabic Handwriting Recognition: A Survey

IEEE Transactions on Pattern Analysis and Machine Intelligence
Finding words in alphabet soup: Inference on freeform character recognition for historical scripts

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we describe four recent enhancements to the BBN Byblos OCR System, a multilingual HMMbased character recognition system which has been demonstrated on a variety of languages, including English, Arabic, Chinese, and Japanese. These enhancements are implemented as optional extensions to the system and provide improved performance for certain scripts or domains. Projection-based reestimation of line boundaries reduces instability in the presence of some types of noise. An alternate modeling strategy used in the first of two recognition search passes substantially increases speed on languages with a large number of characters. Another speed improvement comes from automatic discovery and modeling of sub-characters. The use of Heteroschedastic Linear Discriminant Analysis (HLDA) makes modeling more tractable by reducing feature-space dimensionality.