Research and evaluation of correction methods of geometric distortions of text document images

Author(s) Collection number Pages Download abstract Download full text
Kulchytska I. O., Kulchytskyi R. O., Tymchenko O. O., Tymchenko O. V. № 2 (72) 53-65 Image Image

We have analyzed the existing methods of correction of geometric distortions of text document images and reviewed the weaknesses of each method. The new method of distortion correction has been developed, that does not depend on the type of distortion and can be used to images of a combination of several types of distortions. We have examined the features of algorithms evaluation for correcting distortions of images using OCR. The experimental research has shown that the application of the developed methods for correcting distortions during pre-treatment before the text recognition can significantly improve the recognition quality. The experimental research has shown that the methods of correction of geometric and perspective distortions provide higher quality of pre-treatment levels than commercial software BookRestorer.

Keywords: distorted images, OCR system, text document image, recognition quality, distortion correction, image preprocessing.

  • 1. Zhang, L., Zhang, Y., & Tan, C. L. (2008). An Improved Physically-Based Method for Geometric Restoration of Distorted Document Images. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 30, 4, 728–734 (in English).
  • 2. Ulges, A., Lampert, C. & Breuel, T. M. (2004). Document capture using stereo vision. In Proceedings of the ACM Symposium on Document Engineering. ACM, 198–200 (in English).
  • 3. Yamashita, A., Kawarago, A., Kaneko, T. & Miura, K.T. (2004). Shape reconstruction and image restoration for non-flat surfaces of documents with a stereo vision system. In Proceedings of 17th International Conference on Pattern Recognition (ICPR2004), Vol. 1, 482–485 (in English).
  • 4. Brown, M. S. & Seales, W. B. (2001). Document restoration using 3d shape: A general deskewing algorithm for arbitrarily warped documents. In International Conference on Computer Vision (ICCV01), 2, 367–374 (in English).
  • 5. H. Cao, X. Ding, & C. Liu. (2003). Rectifying the bound document image captured by the camera: A model based approach. 7th International Conference on Document Analysis and Recognition, Scotland, pp. 71–75. (in English).
  • 6. J. Liang, D. DeMenthon, & D. Doermann. (2008). Geometric rectification of camera-captured document images. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 30, 4, pp. 591–605. (in English).
  • 7. C.L. Tan, L. Zhang, Z. Zhang, & T. Xia. (2006). Restoring warped document images through 3D shape modeling. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 28, 2, pp. 195–208. (in English).
  • 8. L. Zhang, A.M. Yip, M.S. Brown and C.L. Tan. (2009). A unified framework for document restoration using inpainting and shape-from-shading. Pattern Recognition, vol. 42, 11, pp. 2961–2978. (in English).
  • 9. A. Masalovich. (2007). Using Bezier patch to approximate the distortion of text documents. Proceedings of the 17th International Conference on Computer Graphics and Vision (GraphiCon 2007), Moscow, pp. 239–243. (in English).
  • 10. Zheng Zhang, Chew Lim Tan. (2003). Correcting Document Image Warping Based on Regression of Curved Text Lines. Proc. ICDAR-2003, pp.589–563. (in English).
  • 11. Brown, Bin, Minghui, Wu, Rongfeng, Li, Wenxin, Li, Zhuoqun, Xu & Chunxu, Yang (2007). A model based book dewarping method using text line detection. Proceedings of the Second International Workshop on Camera-Based Document Analysis and Recognition (CBDAR-2007), 63–70 (in English).
  • 12. Mischke, L. & Luther, W. (2005). Document Image De-warping Based on Detection of Distorted Text Lines. International Conference on Image Analysis and Processing, 1068–1075 (in English).
  • 13. Lavialle, O., Molines, X., Angella, F. & Baylou, P. (2001). Active Contours Network to Straighten Distorted Text Lines. International Conference on Image Processing, 748–751 (in English).
  • 14. Ulges, A., Lampert, C. H. & Breuel, T. M. (2005). Document image dewarping using robust estimation of curled text lines. 8th International Conference on Document Analysis and Recognition, 1001–1005 (in English).
  • 15. Zhang, Y., Liu, C., Ding, X. & Zou, Y. (2008). Arbitrary warped document image restoration based on segmentation and Thin-Plate Splines. 19th International Conference on Pattern Recognition, 1–4 (in English).
  • 16. Kulchytska, I. O., Tymchenko, O. V. & Tymchenko, O. O. (2015). Informatsiina tekhnolohiia vidnovlennia spotvorenykh zobrazhen tekstovykh dokumentiv. Modelling and Information Technologies, 75, 69–79 (in Ukrainian).
  • 17. Kulchytska, I. O., & Tymchenko, O. V. (2013). Osoblyvosti alhorytmiv binaryzatsii zobrazhen dokumentiv. Proceedings of IPM NAS of Ukraine, 68, 141–149 (in Ukrainian).
  • 18. Stamatopoulos, Gatos, B. & Pratikakis, I. (2009). A Methodology for Document Image Dewarping Techniques Performance Evaluation. 10th International Conference on Document Analysis and Recognition, 956–960 (in English).
  • 19. Bin, Fu, Wenxin, Li, Minghui, Wu & Rongfeng, Li (2012). Document Rectification Approach Dealing With Both Perspective Distortion And Warping Based On Text Flow Curve Fitting. International Journal of Image and Graphics, Vol. 12, 1, 23–25 (in English).
  • 20. Kanai, J., Nartker, T. A., Rice, S., & Nagy, G. (1993). Performance metrics for document understanding systems. In Proc. 2nd Int. Conf. Document Anal, 424–427 (in English).
  • 21. Olive, J. (2011). Handbook of Natural Language Processing and Machine Translation: DARPA Global Autonomous Language Exploitation, Springer Science+ Business Media (in English).
  • 22. Levenshtein, V. I. (1966). Binary Codes Capable of Correcting Deletions, Insertions, and Reversals. Soviet Physics Doklady, 10:707–710 (in Enhlish).
  • 23. Datesets University Of Kaiserslautern. Retrieved from: 2009/datasets (in English).
  • 24. OCR Software Review. Retrieved from: (in English).