Control Systems and Computers, N6, 2017, Article 3


Upr. sist. maš., 2017, Issue 6 (272), pp. 26-34, 40.

UDC 004.89:004.93

Volodymyr M. Kyyko  PhD in Techn. Sciences, senior researcher of pattern recognition departmation, International Research and Training Centre of Information Technologies and Systems of the NAS and MES of Ukraine, Glushkov ave., 40, Kyiv, 03187, Ukraine,

License Plate Localization and Recognition in Images  

Introduction. The well known algorithms and systems for license plate localization and recognition are observed. Some ways for increasing robustness of recognition are defined.
Purpose. It is important to provide robust license plates recognition under conditions of long-distance and not frontal location of camera relatively to license plate. To reduce processing time it is desired also preliminary to detect a small number of image parts which may contain license plates and precisely localize these plates.

Methods. Localization of number plates is based on detection of license plate frame contour lines by Hough transform. While recognition special points in symbol contour are detected    and modified Levenstein distances are computed between input and etalon chains of these points. Separation of similar in shape symbols is carried out by detection of additional structural and metrical contour features. Used structural and metrical features of symbols are robust in a great measure to scaling and rotations of symbols in image.

Results. The results of experimental testing of the proposed localization and recognition algorithms are presented. Conclusion. Proposed algorithms can be used as basis for new as well as in developed before systems for license plates recognition in images. 

Keywords: images, license plates localization and recognition, Levenstein distance, Hough transform.

 Download full text!

  1. Automatic number plate recognition. [online] Available at: < recognition> [Accessed 1 Feb. 2017].
  2. DHAWAL, WAZALWAR, ERDAL, ORUKLU, JAFAR, SANIIE, 2012. “A Design Flow for Robust License Plate Localization and Recognition in Complex Scenes”, J. of Transportation Technologies, 2, pp. pp. 13–21.
  3. MARTINSKY, O., 2007. Algorithmic And Mathematical Principles Of Automatic Number Plate Recognition Systems, Bachelor’s thesis, Department of Intelligent Systems, Faculty of Information Technology, Brno University of Technology.
  4. SHIH-HAO, YU, JUN-WEI, HSIEH, YUNG-SHENG, CHEN, 2002. “Morphology-based License Plate Detection from Complex Scenes,” IEEE Proc. Of Int. Conf. on Pattern Recognition, Quebec, Canada (August 11–15, 2002), III, pp. 176–179.
  5. AMR BADR, MOHAMED M. ABDELWAHAB, AHMED M. THABET, AND AHMED M. ABDELSADEK, 2011. “Automatic Number Plate Recognition System”, Annals of the University of Craiova, Mathematics and Computer Science Series, 38(1), pp. 62–71.
  6. MATAS, J., ZIMMERMAN, K., 2005. Unconstrained Licence Plate and Text Localization and Recognition. In: IEEE Int. Conf.on Intell. Transp. Systems, New York, 2005, pp. 225–230.
  7. Duda, R.O., Hart, P.E., 1972. “Use of the Hough Transformation to Detect Lines and Curves in Pictures”, Comm. ACM, 15, pp. 11–15.
  8. KIRYATI, N., ELDAR, Y., BRUCKSTEIN, A.M, 1991. “A Probabilistic Hough transform”, Pattern Recognition, 24, (4), pp. 303–316.
  9. SCHLEZINGER, M.I., 1991. “Fast implementation of one class of linear convolutions”. Theoretical and applied questions of image recognition, Kiev: IK AN USSR, pp. 61–69. (In Russian).
  10. NOBUYUKI, OTSU., 1979. “A threshold selection method from gray-level histograms”. IEEE Trans. Sys., Man., Cyber. 9 (1), pp. 62–66.
  11. LEVENSHTEJN, V.I., 1965. “Binary codes with correction of fallouts, insertions and symbol substitutions”. Reports AN SSSR, M., 163 (4), pp. 845–848. (In Russian).
  12. SHLEZINGER, M.I., VODOLAZSKIY, E.V., YAKOVENKO, V.M., 2014. “Recognition of the similarity of polygons in the strengthened Hausdorff metric”. Cybernetics and Systems Analysis, 3, pp. 174–187. (In Russian).
  13. MURYGIN, K.V., 2010. “Detection of automobile numbers on the basis of the mixed cascade of classifiers”, Artificial Intelligence, 2, pp. 47–152. (In Russian).
  14. MURYGIN, K.V., 2013. “Preliminary processing of candidates when car numbers on images are detected”, Artificial Intelligence, 2, pp. 32–37. (In Russian).
  15. Car license plate number 4. [online] Available at: <> [Accessed 11 Feb. 2017].
  16. SecurOS Auto. [online] Available at: <> [Accessed 12 Feb. 2017].

Received 04.10.2017