Lip Print Recognition Based on Mean Differences Similarity Measure

  • Lukasz Smacki
  • Krzysztof Wrobel
Part of the Advances in Intelligent and Soft Computing book series (AINSC, volume 95)


This paper presents a new method for comparing and classifying lip print images. In the proposed method a lip print image is first binarized and then subjected to the Hough transform. As a result a collection of line segments approximating the lip print pattern is obtained. Each segment is described by its length, angle and midpoint coordinates. Lip prints are compared using the mean differences similarity measure. Presented studies tested the impact of different weights applied to segment’s characteristic features on lip print recognition results. After further improvements the presented method can be used in criminal identification systems.


Feature Extraction Medical Informatics Crime Scene Background Detection Segment Detection 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Campbell, B.M.: Similarity coefficients for classifying releves. Vegetatio 37, 101–109 (1978)CrossRefGoogle Scholar
  2. 2.
    Choras, M.: The lip as a biometric. Pattern Analysis and Applications 13(1), 105–112 (2010)CrossRefGoogle Scholar
  3. 3.
    Czekanowski, J.: Zur differential-diagnose der Neadertalgruppe. Korrespbl. Dt. Ges. Anthrop 40, 44–47 (1909)Google Scholar
  4. 4.
    Duda, R., Hart, P.: Use of the Hough transformation to detect lines and curves in pictures. Comm. ACM 15 (1972)Google Scholar
  5. 5.
    Hough, P.V.C.: Method and means for recognizing complex pattern. U.S. Patent No. 3069654 (1962)Google Scholar
  6. 6.
    Illingworth, J., Kittler, J.: A survey of the Hough transform. Computer Vision Graphics and Image Processing 44(1), 87–116 (1988)CrossRefGoogle Scholar
  7. 7.
    Johnston, J.W.: Similarity Indices I: What Do They Measure?, p. 68 (1976)Google Scholar
  8. 8.
    Kasprzak J., Leczynska B.: Cheiloscopy - Human identification on the basis oflip trace, pp. 11–29 (2001) (in Polish)Google Scholar
  9. 9.
    Newton, M.: The Encyclopedia of Crime Scene Investigation, p. 42 (2008)Google Scholar
  10. 10.
    Porwik, P., Doroz, R., Wrobel, K.: A new signature similarity measure based on windows allocation technique. Int. Journal of Computer Information Systems and Industrial Management Applications (IJCISIM) 2, 297–305 (2010)Google Scholar
  11. 11.
    Sharma, P., Saxwna, S., Rathod, V.: Comparative reliability of cheiloscopy and palatoscopy in human identification. Indian Journal of Dental Research 20(4), 453–457 (2009)CrossRefGoogle Scholar
  12. 12.
    Siegel, J., Saukko, P., Knupfer, G.: Encyclopedia of Forensic Science, pp. 358–362 (2000)Google Scholar
  13. 13.
    Smacki, L.: Lip traces tecognition based on lines pattern. Journal of Medical Informatics and Technologies 15, 53–60 (2010)Google Scholar
  14. 14.
    Smacki, L., Porwik, P., Tomaszycki, K., Kwarcinska, S.: The lip print recognition using Hough transform. Journal of Medical Informatics and Technologies 14, 31–38 (2010)Google Scholar
  15. 15.
    Suzuki, K., Tsuchihashi, Y.: Personal identification by means of lip prints. Journal of Forensic Medicine 17, 52–57 (1970)Google Scholar
  16. 16.
    Tsuchihashi, Y.: Studies on personal identification by means of lip prints. Forensic Science, 127–231 (1974)Google Scholar
  17. 17.
    Wrobel, K., Doroz, R.: New signature similarity measure based on average differences. Journal of Medical Informatics and Technologies 12, 51–56 (2008)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Lukasz Smacki
    • 1
  • Krzysztof Wrobel
    • 1
  1. 1.Institute of Computer ScienceUniversity of SilesiaSosnowiecPoland

Personalised recommendations