Efficient binarization technique for severely degraded document images
Tóm tắt
Degradations in document images appear due to shadows, non-uniform illumination, ink bleed-through and blur caused by humidity. Thresholding of such document images either result in broken characters or detection of false texts. Numerous algorithms exist that can separate text and background efficiently in the textual regions of the document; but portions of background are mistaken as text in areas that hardly contain any text. This paper presents a way to overcome these problems by a robust binarization technique that recovers the text from a severely degraded document images and thereby increases the accuracy of optical character recognition systems. The proposed document recovery algorithm efficiently removes degradations from document images. Proposed work is based on the fusion of two well known binarization methods: Gatos et al. and Niblack, using dilation and logical AND operations. The results of our proposed binarization approach are seen to be better when compared to five existing well known approaches proposed by Otsu, Gatos et al., Niblack, Souvola et al., and Bernsen using four evaluations measures: Execution time, F-measure, PSNR, and NRM.
Tài liệu tham khảo
Lasmar AG, Kricha A, Essoukri N and Amara B (2006) A segmentation text/background method for degraded ancient Arabic manuscript, pp 1327–1331, IEEE
He J, Do QDM, Downton AC and Kim JH (2005) A comparison of binarization methods for historical archive documents. In: Eighth international conference on document analysis and recognition (ICDAR’05), pp 538–542
Jindal M.K, Sharma RK and Lehal GS(2007) A study of different kinds of degradation in printed Gurmukhi script. In: Proceedings of the international conference on computing: theory and applications (ICCTA’07), pp 538–544, IEEE
Otsu N (1979) A threshold selection method from gray-level histograms. IEEE Trans Syst Man Cybernet 9(1):62–66
Trier OD, Jain AK (1995) Goal-directed evaluation of binarization methods. Pattern Anal Mach Intell IEEE Trans 17(12):1191–1201
Gatos B, Pratikakis I, Perantonis SJ (2006) Adaptive degraded document image binarization. Pattern Recognit 39:317–327
Niblack W (1986) An introduction to digital image processing. Prentice-Hall, Englewood Cliffs, pp 115–116
Sauvola J, Pietikainen M (2000) Adaptive document image thresholding. Pattern Recognit 33:225–236
Bernsen J (1986) Dynamic thresholding of grey-level images. In: Proceedings of the eighth ICPR, pp 1251–1255
Kittler J, Illingworth J (1985) On threshold selection using clustering criteria. IEEE Trans Syst Man Cybernet 15:652–655
Brink AD (1992) Thresholding of digital images using two-dimensional entropies. Pattern Recognit 25(8):803–808
Yan H (1996) Unified formulation of a class of image thresholding techniques. Pattern Recognit 29(12):2025–2032
Sahoo PK, Soltani S, Wong AKC (1988) A survey of thresholding techniques. Comput Vis Graph Image Process 41(2):233–260
Kim IK, Jung DW, Park RH (2002) Document image thresholding based on topographic analysis using a water flow model. Pattern Recognit 35:265–277
Yang J, Chen Y, Hsu W (1994) Adaptive thresholding algorithm and its hardware implementation. Pattern Recognit Lett 15(2):141–150
Parker JR, Jennings C and Salkauskas AG (1993) Thresholding using an illumination model. In: International conference on document analysis and recognition (ICDAR’93), pp. 270–273, IEEE
Gonzalez RC, Woods RE, Eddins SL (2010) Digital image processing using MATLAB, 2nd edn. Mc Graw Hill Education, New Delhi, India
The Library of Congress (http://memory.loc.gov/). Accessed Nov 2010
Holy Monastery of St. Catherine at Mount Sinai (http://www.sinaimonastery.com/). Accessed Nov 2010
Bodleian Library—University of Oxford, (http://www.bodley.ox.ac.uk/). Accessed Nov 2010
Users.iit.demokritos.gr/~ bgat/DIBCO2009/benchmark/. Accessed Nov 2010
Users.iit.demokritos.gr/~ bgat/H-DIBCO2010/benchmark/. Accessed Nov 2010
Utopia.duth.gr/~ ipratika/DIBCO2011/benchmark/. Accessed Nov 2010