Advertisement

Pattern Analysis and Applications

, Volume 19, Issue 2, pp 411–426 | Cite as

Arabic character recognition using a Haar cascade classifier approach (HCC)

  • Ashraf AbdelRaoufEmail author
  • Colin A. Higgins
  • Tony Pridmore
  • Mahmoud I. Khalil
Theoretical Advances

Abstract

Optical character recognition (OCR) shows great potential for rapid data entry, but has limited success when applied to the Arabic language. Traditional OCR problems are compounded by the nature of Arabic language and because the script is heavily connected. A machine learning, Haar cascade classifier (HCC) approach was introduced by Viola and Jones (Rapid object detection using a boosted cascade of simple features. Kauai, Hawaii, 2001) to achieve rapid object detection based on a boosted cascade of simple Haar-like features. Here, that approach is applied for the first time to suit Arabic glyph recognition. HCC approach eliminates problematic steps in the pre-processing and recognition phases and, most importantly, character segmentation stage. A classifier was produced for each of the 61 Arabic glyphs that exist after the removal of diacritical marks (dots). These classifiers were trained and tested on some 2,000 images each. The system was tested with real text images and produces a recognition rate for Arabic glyphs of 87 %. The technique gives good results relative to those achieved using a commercial Arabic OCR application and existing state-of-the-art research application.

Keywords

Arabic character recognition Haar-like features Cascade classifiers Haar cascade classifiers 

References

  1. 1.
    Viola P, Jones M (2001) Rapid object detection using a boosted cascade of simple features. In: IEEE conference on computer vision and pattern recognition (CVPR01). Kauai, Hawaii, pp 511–518Google Scholar
  2. 2.
    Abdelazim HY (2006) Recent trends in Arabic character recognition. In: The sixth conference on language engineering. Cairo, Egypt, pp 212–249Google Scholar
  3. 3.
    Adolf F (2003) How-to build a cascade of boosted classifiers based on Haar-like features. OpenCV’s Rapid Object Detection. http://lab.cntl.kyutech.ac.jp/~kobalab/nishida/opencv/OpenCV_ObjectDetection_HowTo.pdf. Accessed 3 Nov 2012
  4. 4.
    Lienhart R, Maydt J (2002) An extended set of Haar-like features for rapid object detection. In: IEEE international conference of image processing (ICIP 2002). New York, USA, pp 900–903Google Scholar
  5. 5.
    Lienhart R, Kuranov A, Pisarevsky V (2002) Empirical analysis of detection cascades of boosted classifiers for rapid object detection. In: 25th pattern recognition symposium (DAGM03). Madgeburg, Germany, pp 297–304Google Scholar
  6. 6.
    Kanoun S, Moalla I, Ennaji A, Alimi AM (2000) Script identification for Arabic and Latin, printed and handwritten documents. Presented at the 4th IAPR-international workshop on document analysis systems: DAS. Rio de Janeiro, BrazilGoogle Scholar
  7. 7.
    Kanoun S, Ennaji A, Lecourtier Y, Alimi AM (2002) Linguistic integration information in the AABATAS Arabic text analysis system. In: 8th international workshop on frontiers in handwriting recognition (IWFHR’02). Ontario, Canada, pp 389–394Google Scholar
  8. 8.
    Kanoun S, Alimi AM, Lecourtier Y (2005) Affixal approach for arabic decomposable vocabulary recognition: a validation on printed word in only one font. In: The 8th international conference on document analysis and recognition (ICDAR’05). Seoul, Korea, pp 1025–1029Google Scholar
  9. 9.
    Kanoun S, Slimane F, Guesmi H, Ingold R, Alimi AM, Hennebert J (2007) Affixal approach versus analytical approach for off-line arabic decomposable vocabulary recognition. In: 10th international conference on document analysis and recognition, ICDAR ‘09. Barcelona, Spain, 661–665Google Scholar
  10. 10.
    Slimane F, Ingold R, Kanoun S, Alimi AM, Hennebert J (2009) A new arabic printed text image database and evaluation protocols. In: 10th international conference on document analysis and recognition. Barcelona, Spain, pp 946–950Google Scholar
  11. 11.
    Benjelil M, Kanoun S, Mullot R, Alimi AM (2010) Complex documents images segmentation based on steerable pyramid features. Int J Doc Anal Recogn 13:209–228CrossRefGoogle Scholar
  12. 12.
    Moussa SB, Zahour A, Benabdelhafid A, Alimi AM (2010) New features using fractal multi-dimensions for generalized Arabic font recognition. Pattern Recogn Lett 31:361–371CrossRefGoogle Scholar
  13. 13.
    Slimane F, Kanoun S, Hennebert J, Alimi AM, Ingold R (2013) A study on font-family and font-size recognition applied to Arabic word images at ultra-low resolution. Pattern Recogn Lett 34:209–218CrossRefGoogle Scholar
  14. 14.
    The Unicode Consortium (2014) The Unicode Standard, Version 7.0.0, The Unicode Consortium, Mountain View, CA. ISBN 978-1-936213-09-2. http://www.unicode.org/versions/Unicode7.0.0/
  15. 15.
    Jaiem FK, Kanoun S, Khemakhem M, El Abed H, Kardoun J (2013) Database for Arabic printed text recognition research. In: ICIAP 2013, Part I, LNCS 8156, pp 251–259Google Scholar
  16. 16.
    AbdelRaouf A, Higgins C, Pridmore T, Khalil M (2010) Building a multi-modal Arabic corpus (MMAC). Int J Doc Anal Recogn 13:285–302CrossRefGoogle Scholar
  17. 17.
    The Unicode Consortium (2011) The Unicode Standard, Version 6.0.0, Chapter 8. The Unicode Consortium, Mountain View, CA. ISBN 978-1-936213-01-6. http://unicode.org/Public/UNIDATA/ArabicShaping.txt. Accessed 11 Apr 2014
  18. 18.
    Ahmed I, Mahmoud SA, Parvez MT (2012) Printed Arabic text recognition. In: Guide to OCR for arabic scripts, Springer, London, pp 147–168Google Scholar
  19. 19.
    Lorigo LM, Govindaraju V (2006) Offline Arabic handwriting recognition: a survey. IEEE Trans Pattern Anal Mach Intell 28:712–724CrossRefGoogle Scholar
  20. 20.
    Amin A (1997) Off line Arabic character recognition—a survey. In: The 4th international conference on document analysis and recognition. Ulm, Germany, pp 596–599Google Scholar
  21. 21.
    Muna LL (2014) Khayyat, Ching Y. Suen, Learning-based word spotting system for Arabic handwritten documents. Pattern Recogn 47:1021–1030CrossRefGoogle Scholar
  22. 22.
    AbdelRaouf A, Higgins C, Khalil M (2008) A database for Arabic printed character recognition. In: The international conference on image analysis and recognition-ICIAR2008. Póvoa de Varzim, Portugal, pp 567–578Google Scholar
  23. 23.
    Alginahi YM (2013) A survey on Arabic character segmentation. Int J Doc Anal Recogn 16:105–126CrossRefGoogle Scholar
  24. 24.
    Harty R, Ghaddar C (2004) Arabic text recognition. Int Arab J Inf Technol 1:156–163Google Scholar
  25. 25.
    Kasinski A, Schmidt A (2010) The architecture and performance of the face and eyes detection system based on the Haar cascade classifiers. Pattern Anal Appl 13:197–211MathSciNetCrossRefGoogle Scholar
  26. 26.
    Crow FC (1984) Summed-area tables for texture mapping. SIGGRAPH Comput Graph 18:207–212CrossRefGoogle Scholar
  27. 27.
    Messom C, Barczak A (2006) Fast and efficient rotated haar-like features using rotated integral images. In: Australian conference on robotics and automation (ACRA2006), pp 1–6Google Scholar
  28. 28.
    AbdelRaouf A, Higgins CA, Pridmore T, Khalil MI (2014) Fast Arabic glyph recognizer based on haar cascade classifiers. Presented at the international conference on pattern recognition applications and methods (ICPRAM 2014). Angers, FranceGoogle Scholar
  29. 29.
    Schapire RE (2002) The boosting approach to machine learning, an overview. In: MSRI workshop on nonlinear estimation and classification. Berkeley, CA, USA, pp 149–172Google Scholar
  30. 30.
    Khorsheed MS (2002) Off-line arabic character recognition—a review. Pattern Anal Appl 5:31–45MathSciNetCrossRefGoogle Scholar
  31. 31.
    Senior A (1992) Off-line handwriting recognition: a review and experiments. Cambridge University, Engineering Department, CambridgeGoogle Scholar
  32. 32.
    Cheriet M, Kharma N, Liu C-L, Suen C (2007) Character recognition systems: a guide for students and practitioners. Wiley, New YorkGoogle Scholar
  33. 33.
    Souza A, Cheriet M, Naoi S, Suen CY (2003) Automatic filter selection using image quality assessment. In: The 7th international conference on document analysis and recognition (ICDAR’03). Edinburgh, ScotlandGoogle Scholar
  34. 34.
    Ahmad I (2013) A technique for skew detection of printed Arabic documents. In: Computer graphics, imaging and visualization (CGIV), 2013 10th international conference, pp 62–67Google Scholar
  35. 35.
    Breuel TM (2002) Robust least square baseline finding using a branch and bound algorithm. In: Document recognition and retrieval VIII, SPIEGoogle Scholar
  36. 36.
    Broumandnia A (2007) Shanbehzadeh J Fast Zernike wavelet moments for Farsi character recognition. Image Vis Comput 25:717–726CrossRefGoogle Scholar
  37. 37.
    Touj S, Amara NEB, Amiri H (2003) Generalized hough transform for Arabic optical character recognition. In: 7th international conference on document analysis and recognition (ICDAR 2003). Edinburgh, Scotland, pp 1242–1246Google Scholar
  38. 38.
    Noor SM, Mohammed IA, George LE (2011) Handwritten Arabic (indian) numerals recognition using Fourier descriptor and structure base classifier. J Al-Nahrain Univ 14:215–224Google Scholar
  39. 39.
    Gonzalez RC, Woods RE (2007) Digital image processing, 3rd edn. Prentice Hall, New Jersey, USAGoogle Scholar
  40. 40.
    Zidouri A (2007) PCA-based Arabic character feature extraction. In: 9th international symposium on signal processing and its applications (ISSPA 2007). Sharjah, United Arab Emirates, pp 1–4Google Scholar
  41. 41.
    Kurt Z, Turkmen HI, Karsligil ME (2009) Linear discriminant analysis in ottoman alphabet character recognition. In: The European computing conference. Tbilisi, Georgia, pp 601–607Google Scholar
  42. 42.
    Trenkle J, Gillies A, Erlandson E, Schlosser S, Cavin S (2001) Advances in Arabic text recognition. In: Symposium on document image understanding technology. Maryland, USAGoogle Scholar
  43. 43.
    Yalniz IZ, Altingovde IS, Güdükbay U, Ulusoy Ö (2009) Integrated segmentation and recognition of connected Ottoman script. Opt Eng 48(11):117205CrossRefGoogle Scholar
  44. 44.
    Sabbour N, Shafait F (2013) A segmentation-free approach to Arabic and Urdu OCR. In: IS&T/SPIE electronic imaging, SPIE digital library, USA, pp 86580 N-86580 N-12Google Scholar
  45. 45.
    Abandah GA, Younis KS, Khedher MZ (2008) Handwritten Arabic character recognition using multiple classifiers based on letter form. In: The 5th iasted international conference on signal processing, pattern recognition and applications (SPPRA 2008). Innsbruck, Austria, pp 128–133Google Scholar
  46. 46.
    Alma’adeed S, Higgens C, Elliman D (2002) Recognition of off-line handwritten Arabic words using hidden markov model approach. In: The 16th international conference on pattern recognition (ICPR’02). Quebec, Canada, pp 481–484Google Scholar
  47. 47.
    Bushofa B, Spann M (1997) Segmentation and recognition of Arabic characters by structural classification. Image Vis Comput 15:167–179CrossRefGoogle Scholar
  48. 48.
    Mehran R, Pirsiavash H, Razzazi F (2005) A front-end OCR for Omni-font Persian/Arabic cursive printed documents. In: Digital image computing: techniques and applications (DICTA’05). Cairns, Australia, pp 56–64Google Scholar
  49. 49.
    Rahman AFR, Fairhurst MC (2003) Multiple classifier decision combination strategies for character recognition: a review. Int J Doc Anal Recogn 5:166–194CrossRefGoogle Scholar
  50. 50.
    OpenCV (2002) Rapid object detection with a cascade of boosted classifiers based on Haar-like features. OpenCV haartraining tutorialGoogle Scholar
  51. 51.
    Sonka M, Hlavac V, Boyle R (1998) Image processing: analysis and machine vision, 2nd edn. Thomson Learning Vocational, Cengage Learning, New Delhi, IndiaGoogle Scholar
  52. 52.
    Box GEP, Muller ME (1958) A note on the generation of random normal deviates. Ann Math Stat 29:610–611CrossRefzbMATHGoogle Scholar
  53. 53.
    Seo N (2008) Tutorial: OpenCV haartraining (rapid object detection with a cascade of boosted classifiers based on Haar-like features)Google Scholar
  54. 54.
    IRIS (2011) Readiris 12 pro. http://www.irislink.com/c2-1684-225/Readiris-12-for-Windows.aspx. Accessed 27 Jul 2011
  55. 55.
    Kohavi R, Provost F (1998) Glossary of terms. special issue on applications of machine learning and the knowledge discovery process. Mach Learn 30:271–274CrossRefGoogle Scholar
  56. 56.
    IRIS (2004) Readiris pro 10, 10th ednGoogle Scholar
  57. 57.
    Schulz KU, Mihov S (2002) Fast string correction with Levenshtein automata. Int J Doc Anal Recogn 5:67–85CrossRefzbMATHGoogle Scholar
  58. 58.
    Garcıa S, Herrera F (2008) An extension on statistical comparisons of classifiers over multiple data sets for all pairwise comparisons. J Mach Learn Res 9:2677–2694zbMATHGoogle Scholar
  59. 59.
    Demsar J (2006) Statistical comparisons of classifiers over multiple data sets. J Mach Learn Res 7:1–30MathSciNetzbMATHGoogle Scholar
  60. 60.
    Webb GI (2000) Multiboosting: a technique for combining boosting and wagging. Mach Learn 40:159–196CrossRefGoogle Scholar
  61. 61.
    Zaiontz C (2013–2015) The data analysis for this paper was generated using the real statistics resource pack software (release 3.5). http://www.real-statistics.com. Accessed 26 Feb 2015

Copyright information

© Springer-Verlag London 2015

Authors and Affiliations

  • Ashraf AbdelRaouf
    • 1
    Email author
  • Colin A. Higgins
    • 2
  • Tony Pridmore
    • 2
  • Mahmoud I. Khalil
    • 3
  1. 1.Faculty of Computer ScienceMisr International UniversityCairoEgypt
  2. 2.School of Computer ScienceThe University of NottinghamNottinghamUK
  3. 3.Computer and Systems Engineering Department, Faculty of EngineeringAin Shams UniversityCairoEgypt

Personalised recommendations