An Evaluation of HMM-Based Techniques for the Recognition of Screen Rendered Text

Segmentation and recognition of screen rendered text is a challenging task due to its low resolution (72 or 96 ppi) and use of antialiased rendering. This paper evaluates Hidden Markov Model (HMM) techniques for OCR of low resolution text -- both on screen rendered isolated characters and screen rendered text-lines -- and compares it with the performance of other commercial and open source OCR systems. Results show that HMM-based methods reach the performance of other methods on screen rendered text and yield above 98% character level accuracies on both screen rendered text-lines and characters.

[1]  Richard M. Schwartz,et al.  An Omnifont Open-Vocabulary OCR System for English and Arabic , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Paul A. Viola,et al.  Text recognition of low-resolution document images , 2005, Eighth International Conference on Document Analysis and Recognition (ICDAR'05).

[3]  Lei Huang,et al.  A video text location method based on background classification , 2009, International Journal on Document Analysis and Recognition (IJDAR).

[4]  Hiroshi Murase,et al.  Recognition of Very Low-Resolution Characters from Motion Images Captured by a Portable Digital Camera , 2004, PCM.

[5]  Steffen Wachenfeld,et al.  Recognition of Screen-Rendered Text , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[6]  Rolf Ingold,et al.  A language-independent, open-vocabulary system based on HMMs for recognition of ultra low resolution words , 2008, SAC '08.

[7]  Alex Waibel,et al.  Readings in speech recognition , 1990 .

[8]  Thomas M. Breuel,et al.  Segmentation of handprinted letter strings using a dynamic programming algorithm , 2001, Proceedings of Sixth International Conference on Document Analysis and Recognition.

[9]  Xiaoyi Jiang,et al.  Annotated Databases for the Recognition of Screen-Rendered Text , 2007, Ninth International Conference on Document Analysis and Recognition (ICDAR 2007).

[10]  Gheith A. Abandah,et al.  Optical character recognition of handwritten Arabic using hidden Markov models , 2011, Defense + Commercial Sensing.

[11]  Mohammad S. Khorsheed,et al.  Offline recognition of omnifont Arabic text using the HMM ToolKit (HTK) , 2007, Pattern Recognit. Lett..

[12]  Vladimir I. Levenshtein,et al.  Binary codes capable of correcting deletions, insertions, and reversals , 1965 .

[13]  Steve Young,et al.  The HTK book version 3.4 , 2006 .

[14]  Horst Bunke,et al.  Using a Statistical Language Model to Improve the Performance of an HMM-Based Cursive Handwriting Recognition System , 2001, Int. J. Pattern Recognit. Artif. Intell..

[15]  Rolf Ingold,et al.  A HMM-Based Approach to Recognize Ultra Low Resolution Anti-Aliased Words , 2007, PReMI.

[16]  Frederick Jelinek,et al.  Statistical methods for speech recognition , 1997 .