Development of modified method for text recognition in standardized picture

Authors

  • Константин Николаевич Касьян Zaporizhzhya National Technical University Zhukovskiy str. 64, Zaporizhzhya, Ukraine, 69063, Ukraine
  • Владимир Владимирович Братчиков Zaporizhzhya National Technical University Zhukovskiy str. 64, Zaporizhzhya, Ukraine, 69063, Ukraine https://orcid.org/0000-0002-6070-7628
  • Вадим Викторович Шкарупило Zaporizhzhya National Technical University Zhukovskoho St64, Zaporizhzhya, Ukraine, 69063, Ukraine https://orcid.org/0000-0002-0523-8910

DOI:

https://doi.org/10.15587/1729-4061.2015.43047

Keywords:

text recognition, template method, standard, neural network, perceptron, license plate, OpenCV

Abstract

Text recognition in images is a very urgent problem in modern search engines. There are many different methods and techniques for text recognition. The paper is a method for text recognition in a standardized image. Standardized image means an image that has the same font, character size, certain writing order, such as the serial number or license plate of the car.

In the paper, we developed an improved method for text recognition in the image. The method consists in a preliminary search of the same characters and memorizing their positions. Identical symbols are recognized only once. After recognition, symbols are arranged in the desired position. Image processing and isolation of character boundaries is performed using JavaCV.

The modified method was developed based on the template method. Both methods were implemented in Java language. To create a text-recognition software, a neural network based on a single-layer perceptron was built. The results of tests have shown the superiority of the modified method compared to the original one. At best, the performance of the modified method is 300% of the performance of the original one. At worst, it is slower only by 5-10%. In addition, the modified algorithm requires 3 times fewer iterations.

The modified algorithm allows to accelerate the text recognition process in standardized images if they have recurring characters.

Author Biographies

Константин Николаевич Касьян, Zaporizhzhya National Technical University Zhukovskiy str. 64, Zaporizhzhya, Ukraine, 69063

Ph.D., Associate Professor

Department of Computer Systems and Networks

Владимир Владимирович Братчиков, Zaporizhzhya National Technical University Zhukovskiy str. 64, Zaporizhzhya, Ukraine, 69063

Department of Computer Systems and Networks

Вадим Викторович Шкарупило, Zaporizhzhya National Technical University Zhukovskoho St64, Zaporizhzhya, Ukraine, 69063

PhD

Computer Systems and Networks Department

References

  1. Syuzev, V. (2012). Hybrid method OCR correction of recognition results. Engineering. Journal: science and innovation, 11, 12.
  2. Nguyen, Thi Khanh Tien (2014). Detection and recognition of texts in images of complex graphics scenes by using convolutional neural networks. Electrical and computer systems, 13, 125–130.
  3. Mokshin, V., Gabdrakhmanova, L. (2014). Developing character recognition system using neural network. Modern innovations in science and technology, 4, 223–225.
  4. Zelencov, I., Filipovic, Y. (2011). Pattern recognition based on structural frame-based descriptions in shorthand texts XVII century. Science and education: electronic science and technology publication, 12, 28.
  5. Kuchuganov, A., Shards, P. (2008). Recognition of Old Church Slavonic texts methods based on bioalgoritmah image analysis. Modern Information Technologies and Written Heritage: From ancient texts to electronic libraries. el'manuscript-08, 168–172.
  6. Kubrin, S., Mabuza, N., Isaev, A., Masters, S. (2005). Comparison of the geometric moments and Fourier descriptors method in problems of OCR. Mountain information-analytical bulletin (scientific and technical journal), 3, 106–108.
  7. Phan, Ngoc, Bui, Thi Thu Chang1, Spitcin, V. (2012). Hoang. Recognition of printed texts by applying the wavelet transform and principal component analysis. Bulletin of the Tomsk Polytechnic University, 5, 154–157.
  8. Yan, J., Gao, X. (2014). Detection and recognition of text superimposed in images base on layered method. Neurocomputing, 134, 3–14. doi: 10.1016/j.neucom.2012.12.070
  9. González, Á., Bergasa, L. M. (2013). A text reading algorithm for natural images. Image and Vision Computing, 31 (3), 255–274. doi: 10.1016/j.imavis.2013.01.003
  10. Ubozhenko, N. (2013). Analysis of the effectiveness of methods of character recognition tasks as part of the license plate recognition vehicle. Prospects of development of information technologies, 12, 41–45.
  11. Kachanovsky, Y., Yavtuhovich, A. (2007). Development license plate localization algorithm for use in a distributed hardware-software complex ANPR. Information Technology modeling and management, 39, 508–516.
  12. Petrov, S. (2013). Convolutional neural network for character recognition license plate of the car. System analysis in science and education, 21, 66–73.
  13. Gorban, A., Dunin-Barkovskii, V., Kardin, A. (1998). Neuroinformatics. RAS, Sib. Dep., Institute of calc. Simulation, 296.
  14. Uosserman, F. (1992). Neurocomputing equipment: Theory and Practice. Mir, 184.
  15. Fedotov, N. (1990). Methods of stochastic geometry in pattern recognition. Radio and Communications, 144.
  16. Yaser, S. (2012). Learning From Data. AMLBook, 213.
  17. Flanagan, C. (2013). OCR Psychology: AS Revision. Psychology Press, 88. doi: 10.4324/9780203796665
  18. Parker, J. R. (2010). Algorithms for Image Processing and Computer Vision. Wiley, 504.
  19. Bloch, J. (2008). Effective Java 2nd Edition. Sun Microsystems, Inc, Santa Clara, California 95054 USA, 369.
  20. Hominchenko, D. (2013). Configuring JavaCV for windows. Brest. Available at: http://habrahabr.ru/post/190104
  21. OpenCV – Documentation. Available at: http://docs.opencv.org/
  22. Bradski, G. R., Kaehler, A. (2011). Learning OpenCV. O'Reilly Media, Inc., 1005 Gravenstein Highway North, Sebastopol, CA 95472, 571.
  23. Forsyth, D. A., Ponce, J. (2011). Computer Vision: A Modern. Pearson, 792.
  24. Ubozhenko, N. Analysis methods for image pre-processing in the framework of the recognition problem of dirty and / or noisy license plates of vehicles. Prospects of development of information technologies, 18, 57–61.

Downloads

Published

2015-06-29

How to Cite

Касьян, К. Н., Братчиков, В. В., & Шкарупило, В. В. (2015). Development of modified method for text recognition in standardized picture. Eastern-European Journal of Enterprise Technologies, 3(2(75), 11–17. https://doi.org/10.15587/1729-4061.2015.43047

Issue

Vol. 3 No. 2(75) (2015): Information technology. Industry control systems

Section

Information technology

License

Copyright (c) 2015 Константин Николаевич Касьян, Владимир Владимирович Братчиков, Вадим Викторович Шкарупило

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

The consolidation and conditions for the transfer of copyright (identification of authorship) is carried out in the License Agreement. In particular, the authors reserve the right to the authorship of their manuscript and transfer the first publication of this work to the journal under the terms of the Creative Commons CC BY license. At the same time, they have the right to conclude on their own additional agreements concerning the non-exclusive distribution of the work in the form in which it was published by this journal, but provided that the link to the first publication of the article in this journal is preserved.

A license agreement is a document in which the author warrants that he/she owns all copyright for the work (manuscript, article, etc.).
The authors, signing the License Agreement with TECHNOLOGY CENTER PC, have all rights to the further use of their work, provided that they link to our edition in which the work was published.
According to the terms of the License Agreement, the Publisher TECHNOLOGY CENTER PC does not take away your copyrights and receives permission from the authors to use and dissemination of the publication through the world's scientific resources (own electronic resources, scientometric databases, repositories, libraries, etc.).
In the absence of a signed License Agreement or in the absence of this agreement of identifiers allowing to identify the identity of the author, the editors have no right to work with the manuscript.
It is important to remember that there is another type of agreement between authors and publishers – when copyright is transferred from the authors to the publisher. In this case, the authors lose ownership of their work and may not use it in any way.