Road sign recognition using support vector machines and histogram of oriented gradients
S.O. Lisitsyn, O.A. Bayda

Full text of article: Russian language.

Abstract:
In this paper, we consider the recognition of traffic signs using support vector machines (SVMs) and features based on histograms of oriented gradients (HOG). We approach the training of classifier with two well developed multiclass support vector machine formulations proposed earlier by Weston & Watkins and Crammer & Singer. Feature space straightening is approached with Jensen-Shannon and histogram intersection kernels. Due to computational efficiency reasons we propose the use of the homogeneous kernel mapping presented recently. The comparative study based on the German Road Traffic Sign Recognition Benchmark dataset shows the effectiveness of our approach.

Key words:
machine learning, pattern recognition, histogram of oriented gradients, multiclass support vector machines.

References:

  1. Lowe, D.G. Object recognition from local scale-invariant features / D.G. Lowe // Proceedings of the Seventh IEEE International Conference on Computer Vision. – 1999. – Vol. 2. – P. 1150-1157.
  2. Ojala, T. Performance evaluation of texture measures with classification based on Kullback discrimination of distributions / T. Ojala, M. Pietikainen, D. Harwood // Proceedings of the 12th IAPR International Conference. – 1994. – Vol. 1 – P. 582-585.
  3. Dalal, N. Histograms of Oriented Gradients for Human Detection. / N. Dalal, W. Triggs // IEEE Computer Society Conference on Computer Vision and Pattern Recognition CVPR05. – 2005. – Vol. 1(3). – P. 886-893.
  4. Ke, Y. PCA-SIFT : A More Distinctive Representation for Local Image Descriptors / Y. Ke, R. Sikthankar. – Computer Vision and Pattern Recognition, 2004.
  5. Vapnik, V.N. Pattern recognition theory / V.N. Vapnik, A.Ya. Chervonenkis // Moscow: “Nauka” Publisher, 1974. – 416 p. – (In Russian).
  6. Wang, X. An HOG-LBP Human Detector with Partial Occlusion Handling / X. Wang, T.X. Han // IEEE Computer Engineering. – 2009. – P. 32-39.
  7. Ruta, A. In-vehicle camera traffic sign detection and recognition / A. Ruta, F. Porikli, S. Watanabe, Y. Li // Machine Vision and Applications. – 2011. – Vol. 22(2) – P. 359-375.
  8. Shapiro, L. Computer vision / L. Shapiro, G. Stockman // Prentice Hall, 2001. – 608 p.
  9. Bordes, A. Solving multiclass support vector machines with LaRank / A. Bordes, L. Bottou, P. Gallinari, J. Weston // Proceedings of the 24th International Conference on Machine Learning. – 2007. – P. 89-96.
  10. Franc, V. Multi-class Support Vector Machine / V. Franc, V. Hlavac // IEEE Communication. – 2002. – Vol. 2(1) – P. 5-8.
  11. Mitchell, B.F. Finding point of polyhedron nearest to the origin / B.F. Mitchell, V.F. Demyanov, V.N. Malozemov // Leningrad: “Vestnik LGU”. – 1971. – N 19 – P. 38-45 – (In Russian).
  12. Crammer, K. On the Algorithmic Implementation of Multiclass Kernel-based Vector Machines / K. Crammer, Y. Singer // Journal of Machine Learning Research. – Vol. 2. – P. 265-292.
  13. Keerthi, S.S. A sequential dual method for large scale multi-class linear SVMs / S.S. Keerthi, S. Sundararajan, K.-W. Chang, C.-J. Hsieh, C.-J. Lin // Proceeding of the 14th ACM SIGKDD international conference on Knowedge discovery and data mining KDD. – 2008. – P. 408-413.
  14. Aizerman, M.A. Theoretical foundations of the potential function method in pattern recognition learning / M.A. Ai­zerman, M.A. Braverman, L.I. Rosonoer // Moscow: “Na­uka” Publisher, 1970. – 384 p. – (In Russian).
  15. Barla, A. Histogram intersection kernel for image classification. / A. Barla, F. Odone, A. Verri // Proceedings of International Conference on Image Processing. – 2003. – Vol. 2(1). – P. 513.
  16. Martins, A.F.T. Nonextensive entropic kernels / A.F.T. Mar­tins, M.A.T. Figueiredo, P.M.Q. Aguiar, N.A. Smith, E.P. Xing // Proceedings of the 25th International Conference on Machine Learning. – 2008. – P. 640-647.
  17. Maji, S. Classification using intersection kernel support vector machines is efficient / S. Maji, A.C. Berg, J. Malik // IEEE Conference on Computer Vision and Pattern Recognition. – 2008. – P. 1-8.
  18. Vedaldi, A. Efficient Additive Kernels via Explicit Feature Maps / A. Vedaldi, A. Zisserman // IEEE Engineering. – 2011. – Vol. 10(3) – P. 3539-3546.
  19. Yosida, K. Functional analysis / K. Yosida // Springer classic in mathematics, 1996. – 513 p.
  20. Sonnenburg, S. The SHOGUN Machine Learning Toolbox / S. Sonnenburg, G. Raetsch, S. Henschel, C. Widmer, J. Behr, A. Zien, F. De Bona // Journal of Machine Learning Research. – 2010. – Vol. 11(5). – P. 1799-1802.
© 2009, IPSI RAS
Institution of Russian Academy of Sciences, Image Processing Systems Institute of RAS, Russia, 443001, Samara, Molodogvardeyskaya Street 151; E-mail: ko@smr.ru; Phones: +7 (846 2) 332-56-22, Fax: +7 (846 2) 332-56-20