Abstract
Given font/type-written text rectangle bitmaps extracted from digitally scanned pages, inferring the boundaries of lines hence complete words is a preprocessing vital to whatever OCR system while the recognition process itself as well as the post processing necessary for producing the recognized text.
Histogram-based methods are commonly used due mainly to their relative implementation simplicity and computational efficiency, however, some authors report about their vulnerability to some idiosyncratic textual structure complexities, and noise.
This paper elaborates on this approach to produce a more robust algorithm for lines/words decomposition esp. in Arabic, or Arabic dominated, text rectangles from real-life multifont/multisize documents. Trying to evaluate this algorithm, this paper also presents the results of extensive experimentation made on about 1800 documents fairly distributed over different kinds of sources with different noise levels at different scanning resolutions and color depths.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Amin, A., Masini, G.: Machine recognition of multi-font printed Arabic texts. In: Proc. 8th International Joint Conf. on Pattern Recognition, Paris, France, pp. 392-395 (October 1986)
Al-Badr, B., Mahmoud, S.A.: Survey and Bibliography of Arabic Optical Text Recognition. Signal Processing 41, 49–77 (1995)
Attia, M., El-Mahallawy, M.: Histogram-Based Decomposition Algorithm Evaluation DB, http://www.RDI-eg.com/OCR_Decomp_Eval_DB
Attia, M.: Arabic Orthography vs. Arabic OCR. Multilingual Computing & Technology magazine, USA (December 2004)
Bazzi, I., Schwartz, R., Makhoul, J.: An Omni font Open-Vocabulary OCR System for English and Arabic. IEEE Transactions on Pattern Analysis and Machine Intelligence 21(6) (1999)
Bouhilali, K., Kamrouni, M., Ellouze, N.: Method of segmentation of Arabic text image into characters. In: Proceedings of the 1st Kuwaiti Computer Conf., Kuwait, pp.442–446 (March 1989)
Fujisawa, H., Nakano, Y., Kurino, K.: Segmentation Methods for Character Recognition: From Segmentation to Document Structure Analysis. Proc. IEEE 80(7), 1079–1091 (1992)
Gonzalez, R., Woods, R.: Digital Image Processing, 2nd edn. Prentice-Hall, Englewood Cliffs (2002)
Gouda, A.M.: Arabic Handwritten Connected Character Recognition, PhD thesis, Dept. of Computer Engineering, Faculty of Engineering, Cairo University (November 2004)
Kasturi, R., O’Gorman, L.: Document image analysis: A bibliography. Machine Vision and Appl. 5, 231–243 (1992)
Parhami, B., Taraghi, M.: Automatic recognition of printed Farsi texts. Pattern Recognition 14(1), 1–6 (1981)
Pratt, W.K.: Digital Image Processing, 2nd edn. John Wiley & Sons Inc., West Sussex, England (1991)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Attia, M., El-Mahallawy, M. (2007). Histogram-Based Lines and Words Decomposition for Arabic Omni Font-Written OCR Systems; Enhancements and Evaluation. In: Kropatsch, W.G., Kampel, M., Hanbury, A. (eds) Computer Analysis of Images and Patterns. CAIP 2007. Lecture Notes in Computer Science, vol 4673. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-74272-2_65
Download citation
DOI: https://doi.org/10.1007/978-3-540-74272-2_65
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-74271-5
Online ISBN: 978-3-540-74272-2
eBook Packages: Computer ScienceComputer Science (R0)