MAPS: midline analysis and propagation of segmentation

ABSTRACT

Scenic word images undergo degradations due to motion blur, uneven illumination, shadows and defocussing, which lead to difficulty in segmentation. As a result, the recognition results reported on the scenic word image datasets of ICDAR have been low. We introduce a novel technique, where we choose the middle row of the image as a sub-image and segment it first. Then, the labels from this segmented sub-image are used to propagate labels to other pixels in the image. This approach, which is unique and distinct from the existing methods, results in improved segmentation. Bayesian classification and Max-flow methods have been independently used for label propagation. This midline based approach limits the impact of degradations that happens to the image. The segmented text image is recognized using the trial version of Omnipage OCR. We have tested our method on ICDAR 2003 and ICDAR 2011 datasets. Our word recognition results of 64.5% and 71.6% are better than those of methods in the literature and also methods that competed in the Robust reading competition. Our method makes an implicit assumption that degradation is not present in the middle row.

References

1. N. Otsu, A Thresholding Selection Method from Gray-level Histogram, IEEE Transanctions on Systems, Man and Cybernetics, vol. 9, pp. 62--66, March 1979.
2. J. Kittler, J. Illingworth, and J. Foglein, Threshold selection based on a simple image statistic, Computer Vision, Graphics, and Image Processing, vol. 30, no. 2, pp. 125--147, 1985.
3. J. Canny, A Computational Approach to Edge Detection, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 8, no. 6, pp. 679--698, November 1986.
4. W. Niblack, An introduction to digital image processing. New York: Prentice Hall, 1986.
5. J. J. Sauvola and M. Pietäikinen, Adaptive document image binarization, Pattern Recognition, vol. 33, no. 2, pp. 225--236, 2000.
6. R. O. Duda, P. E. Hart and D. G. Stork., Pattern classification, Wiley, 2001.
7. H. C. Thode, Jr., Testing for Normality, New York, Marcel Dekker, 2002.
8. J. Matas, O. Chum, M. Urban and T. Pajdla, Robust wide baseline stereo from maximally stable extremal regions, British Machine Vision Conference, pp. 384--393, 2002.
9. Y. Boykov and V. Kolmogorov, An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Energy Minimization in Vision, IEEE Trans. PAMI, vol. 26, no. 9, pp. 1124--1137, September 2004.
10. S. M. Lucas et.al, ICDAR 2003 Robust Reading Competitions: Entries, Results, and Future Directions, International Journal on Document Analysis and Recognition, vol. 7, no. 2, pp. 105--122, June 2005.
11. S. M. Lucas, Text Locating Competition Results, Proc. 8th Intl. Conf. on Document Analysis and Recognition (ICDAR) 2005, pp. 80--85, 2005.
12. H. Liu, X. Ding, Handwritten Character Recognition Using Gradient Feature and Quadratic Classifier with Multiple Discrimination Schemes, Proc. 8th Int. Conf. on Document Analysis And Recognition, pp. 19--25, 2005.
13. C. M. Thillou and B. Gosselin, Color text extraction from camera-captured images: the impact of the choice of the clustering distance, Proc. of 8th Int. Conf. on Document Analysis And Recognition, pp. 312--316, 2005.
14. C. M. Thillou and B. Gosselin, Color text extraction with selective metric-based clustering, Computer, Vision and Image Understanding, vol. 107, no. 2, pp. 97--107, 2007.
15. T. Kasar, J. Kumar and A. G. Ramakrishnan, Font and background color independent text binarization, Proc 2nd Camera-based Document Analysis and Recognition (CBDAR), pp. 3--9, 2007.
16. T. Kasar and A. G. Ramakrishnan, COCOCLUST: Contour-based color clustering for robust binarization of colored text, Proc 3rd Camera-based Document Analysis and Recognition (CBDAR), pp. 11--17, 2009.
17. L. Neumann and J. Matas, A Method for Text Localization and Recognition in Real-World Images, Proc. 10th Asian Conference on Computer Vision (ACCV), pp. 770--783, 2010.
18. C. Zeng, W. Jia and X. He, An Algorithm for Colour-based Natural Scene Text Segmentation, Proc 4th Camera-based Document Analysis and Recognition (CBDAR), pp. 67--72, 2011.
19. A. Mishra, K. Alahari and C. V. Jawahar, An MRF Model for Binarization of Natural Scene Text, Proc. 11th International Conference of Document Analysis and Recognition, pp. 11--16, September 2011.
20. A. Shahab, F. Shafait and A. Dengel, ICDAR 2011 Robust Reading Competition - Challenge 2: Reading Text in Scene Images, In Proc. 11th International Conference of Document Analysis and Recognition, pp. 1491--1496, September 2011.
21. K. Wang, B. Babenko and S. Belongie, End-to-End Scene Text Recognition, Proc. 13th International Conference on Computer Vision (ICCV), pp. 1457--1464, 2011.
22. A. Mishra, K. Alahari and C. V. Jawahar, Top-Down and Bottom-Up Cues for Scene Text Recognition, Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2012.
23. D. Kumar, M. N. Anil Prasad and A. G. Ramkrishnan, Benchmarking recognition results on word image datasets, CoRR, vol. abs/1208.6137, 2012. http://arxiv.org/abs/1208.6137
24. "Abbyy Fine reader," http://www.abbyy.com/.
25. "Adobe Reader," http://www.adobe.com/products/acrobatpro/scanning-ocr-to-pdf.html.
26. IAPR TC11 Reading Systems-Datasets List, http://www.iapr-tc11/mediawiki/index.php/Datasets
27. "Inzisoft," http://www.inzisoft.com/english/.
28. "KAIST AIPR," http://ai.kaist.ac.kr/home/.
29. "Nuance Omnipage reader," http://www.nuance.com/.
30. "Tesseract OCR engine," http://code.google.com/p/tesseract-ocr/.

Index Terms

• Computing methodologies

  • Artificial intelligence

    • Computer vision

      • Computer vision problems

        • Image segmentation

        • Video segmentation

      • Computer vision tasks