Improvements in hidden Markov model based Arabic OCR

This paper describes recent advances in hidden Markov model (HMM) based OCR for machine-printed Arabic documents. A combination of scriptindependent and script-specific techniques are applied to glyph models and language models (LM). Scriptindependent techniques we applied are higher order ngram LMs for N-best rescoring and discriminative estimation of glyph HMMs. Arabic specific techniques include the use of context-dependent HMMs for glyph modeling and Parts-of-Arabic-Words in language modeling. We present experimental results that demonstrate a 40% relative reduction in word error rate over the baseline configuration on a corpus of machine-printed Arabic documents.

• Type: Conference
• Year: 2008
• Where: ICPR
• Authors: Rohit Prasad, Shirin Saleem, Matin Kamali, Ralf Meermeier, Premkumar Natarajan
• Keywords: Arabic Specific Techniques, Computer Vision, Glyph Hmms, ICPR 2008, Machine-printed Arabic Documents