Radiotekhnika
Publishing house Radiotekhnika

"Publishing house Radiotekhnika":
scientific and technical literature.
Books and journals of publishing houses: IPRZHR, RS-PRESS, SCIENCE-PRESS


Тел.: +7 (495) 625-9241

 

Development and analysis of face recognition algorithm based on local quantized patterns

Keywords:

A. Nikitin – Post-graduate Student, P.G. Demidov Yaroslavl State University. E-mail: anatolynikitinyar@gmail.com
V. Khryashchev – Ph.D. (Eng.), Associate Professor, P.G. Demidov Yaroslavl State University. E-mail: vhr@yandex.ru
A. Priorov – Dr.Sc. (Eng.), Associate Professor, P.G. Demidov Yaroslavl State University. E-mail: andcat@yandex.ru
D. Matveevy – Post-graduate Student, P.G. Demidov Yaroslavl State University. E-mail: dcslab@uniyar.ac.ru


Face recognition is an actual task which has many potential applications for automatically identifying or verifying a person from a digital image or from a video data in security systems. Face recognition process consist of two stages: face detection and face identification. The modern face recognition algorithms extract facial features from an image and compare them with features from the images saved in database. Up to now, many algorithms have been applied for face recognition. They were based on geometric features extraction, facial keypoints detection, Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Local Binary Patterns (LBP), a set of Gabor filters, Discrete Cosine Transform (DCT) etc. Due to its discriminative power and computational simplicity LBP methodology has become one of the most popular approaches for texture classification and face recognition. Image noise might significantly lower accuracy of face detection system based on simple 8-bit LBP features. Unfortunately, we cannot simple expand the neighborhood of each pixel which uses for computing LBP, because the code size increases exponentially with the number of neighborhood pixels. That is why we introduce Local Quantized Patterns (LQP) uses vocabulary to translate a long length binary pattern into a label of the cluster which includes this pattern. We use k-means clustering as a method of dividing all probable binary patterns into k clusters and creating vocabulary. We divide the face image into cells. Therefore over the cell we compute the histogram of the frequency of each «label of the cluster» occurring for each pixel. Then we concatenate histograms of all cells and get the feature vector for the face image. We compared and contrasted the proposed algorithm based on LQP with algorithms based on LBP, PCA and LDA. Comparative study on the AT&T Face Database and our own Russian persons database showed that the developed algorithm can reduce the number of recognition errors by 1.5-2 times compared to similar algorithm. Trained recognition algorithm based LQP able to work with video in real time. The main disadvantage of the proposed algorithm is significantly more time needed for training compared with the considered approach.
References:

  1. Zhao W., Chellappa R., Phillips P., Rosenfeld A. Face recognition: A literature survey // ACM Computing Surveys (CSUR). 2003. V. 35. № 4. P. 399-458.
  2. Szeliski R. Computer Vision: Algorithms and Applications. Springer. 2010. 832 p.
  3. Priorov A.L., Ignatov I.S., Golubev M.N., Xryashhev V.V. Razrabotka i analiz algoritmov vy'deleniya licz na izobrazheniyax // Proektirovanie i texnologiya e'lektronny'x sredstv. 2008. № 2. S. 58-62.
  4. Shmaglit L.A., Priorov A.L., Xryashhev V.V., Matveev D.V. Detektirovanie licz na izobrazheniyax v usloviyax additivnogo belogo gaussovskogo shuma // E'lektromagnitny'e volny' i e'lektronny'e sistemy'. 2014. T. 19. № 5. S. 62–70.
  5. Zhuravlev Ju.I. Ob algebraicheskom podxode k resheniyu zadach raspoznavaniya i klassifikaczii // Problemy' kibernetiki. M.: Nauka. 1978. № 33. S. 5-68.
  6. Kanade T. Picture processing system by computer complex and recognition of human faces. PhD thesis, Kyoto University. 1973.
  7. Brunelli R., Poggio T. Face recognition through Geometrical Features // European Conference on Computer Vision (ECCV). 1992. P. 792–800.
  8. Turk M., Pentland A. Eigenfaces for recognition // Journal of Cognitive Neuroscience. № 3. 1991. P. 71–86.
  9. Belhumeur P.N., Hespanha J., Kriegman D. Eigenfaces vs Fisherfaces: recognition using class specific linear projection // IEEE Transactions on Pattern Analysis and Machine Intelligence. 1997. V. 19. P. 711–720.
  10. Wiskott L., Fellous J., Krüger N., Malsburg C. Face recognition by elastic bunch graph matching // IEEE Transactions on Pattern Analysis and Machine Intelligence. № 3. V. 19. 1997. P. 775–779.
  11. Messer K., Kittler J., Short J. Performance characterization of face recognition algorithms and their sensitivity to severe illumination changes // International Conference on Biometrics (ICB). 2006. P. 1–11.
  12. Ojala T., Pietikäinen M., Harwood D. Performance evaluation of texture measures with classification based on Kullback discrimination of distributions // 12th IAPR International Conference on Pattern Recognition (ICPR). 1994. V. 1. P. 582–585.
  13. Ahonen T., Hadid A., Pietikäinen M. Face recognition with local binary patterns // European Conference on Computer Vision (ECCV). 2004. P. 469–481.
  14. Huang D., Shan C., Ardabilian M., Wang Y., Chen L. Local binary patterns and its application to facial image analysis: a survey // IEEE transactions on systems, man, and cybernetics – part C: applications and reviews. 2011. V. 41. № 6. P. 765–781.
  15. Petruk V.I., Samorodov A.V., Spiridonov I.N. Primenenie lokal'ny'x binarny'x shablonov k resheniyu zadachi raspoznavaniya licz // Vestnik Moskovskogo gosudarstvennogo texnicheskogo universiteta im. N.E'. Baumana. Ser. Priborostroenie. 2011. Specz. vy'p. Biometricheskie texnologii. S. 58–63.
  16. Hussain S., Triggs B. Visual recognition using Local Quantized Patterns // European Conference on Computer Vision (ECCV). 2012. P. 716–729.
  17. Mandel' I.D. Klasterny'j analiz. M.: Finansy' i statistika. 1988. 176 s.
  18. MacQueen J.B. Some methods for classification and analysis of multivariate observations // 5th Berkeley Symposium on mathematical statistics and probability. University of California Press. 1967. P. 281–297.
  19. Hsu C.-W., Lin C.-J. A comparison of methods for multi-class support vector machines // IEEE Transactions on Neural Networks. 2002. V. 13. P. 415–425.
  20. Duan K.B., Keerthi S.S. Which is the best Multiclass SVM method? An empirical study // Multiple Classifier Systems. Lecture Notes in Computer Science 3541. 2005. P. 278–285.
  21. The Database of Faces, AT&T Laboratories Cambridge. http://www.cl.cam.ac.uk/research/dtg/attarchive/facedatabase.html.
  22. Open Source Computer Vision Library (OpenCV), http://opencv.org.

June 24, 2020
May 29, 2020

© Издательство «РАДИОТЕХНИКА», 2004-2017            Тел.: (495) 625-9241                   Designed by [SWAP]Studio