DoCEIS 2010: Emerging Trends in Technological Innovation pp 324-331 | Cite as
An Approach to Modification of Water Flow Algorithm for Segmentation and Text Parameters Extraction
Abstract
This paper proposes an approach to water flow method modification for text segmentation and reference text line detection of sample text at almost any skew angle. Original water flow algorithm assumes hypothetical water flows under only a few specified angles of the document image frame from left to right and vice versa. As a result of water flow algorithm, unwetted image frames are extracted. These areas are of major importance for text line parameters extraction as well as for text segmentation. Water flow method modification means extension values of water flow specified angle and unwetted image frames function enlargement. Modified method is examined and evaluated under different sample text skew angles. Results are encouraged especially due to improving text segmentation which is the most challenging process stage.
Keywords
Document image processing Reference text line Text line segmentation Water flow algorithmReferences
- 1.Basu, S., Chaudhuri, C., Kundu, M., Nasipuri, M., Basu, D.K.: Text Line Extraction from Multi-Skewed Handwritten Documents. Pattern Recognition 40(6), 1825–1829 (2006)CrossRefMATHGoogle Scholar
- 2.Amin, A., Wu, S.: Robust Skew Detection in mixed Text/Graphics Documents. In: Proceedings of 8th International Conference on Document Analysis and Recognition (ICDAR 2005), Seoul, Korea, pp. 247–251 (2005)Google Scholar
- 3.Louloudis, G., Gatos, B., Pratikakis, I., Halatsis, C.: Text Line Detection in Handwritten Documents. Pattern Recognition 41(12), 3758–3772 (2008)CrossRefMATHGoogle Scholar
- 4.Li, Y., Zheng, Y., Doermann, D., Jaeger, S.: A New Algorithm for Detecting Text Line in Handwritten Documents. In: Proceedings of 18th International Conference on Pattern Recognition, Hong Kong, China, vol. 2, pp. 1030–1033 (2006)Google Scholar
- 5.Wang, J., Mazlor, K., Leung, H., Hui, S.C.: Cursive Word Reference Line Detection. Pattern Recognition 30(3), 503–511 (1997)CrossRefGoogle Scholar
- 6.Brodić, D., Milivojević, Z.: Reference Text Line Identification Based on Water Flow Algorithm. In: Proceedings of ICEST 2009, SP-2 Sect, Veliko Tarnovo, Bulgaria (2009)Google Scholar
- 7.Brodić, D., Milivojević, Z.: Modified Water Flow Method for Reference Text Line Detection. In: Proceedings of ICCS 2009, Sofia, Bulgaria (2009)Google Scholar
- 8.Gonzalez, R.C., Woods, R.E.: Digital Image Procesing, 2nd edn., pp. 67–70. Prentice-Hall, New Jersey (2002)Google Scholar
- 9.Sonka, M., Hlavac, V., Boyle, R.: Image Processing, Analysis and Machine Vision, pp. 174–177. Thomson, Toronto (2008)Google Scholar
- 10.Bolstad, W.M.: Introduction to Bayesian Statistics, pp. 40–44, 235–240. John Wiley & Sons, New Jersey (2004)Google Scholar