Descriptor Combination using Firefly Algorithm - Iris.sel.eesc.sc.usp.br

difference of over 2% for Corel database and almost 1% forFree Photo database. The low gain might be caused due tothe usage of a descriptor that has low performance in suchconfiguration of datasets.VII. CONCLUSIONSMany approaches have proposed feature combination in imageclassification. This paper has presented techniques basedon social dynamics to perform descriptor combination usingoptimization algorithms and Firefly Algorithm is introduced inthis context, which is defined by a pair of feature extractionalgorithm and a distance function associated with it.The approach consists in a mathematical formulation of acomposite descriptor, which is obtained by combining descriptorsusing PSO, HS and FFA, and the OPF to perform the tests.In this case, the accuracy rate of OPF in a validation set is usedas an objective function to be maximized by these techniques.Experiments with color and texture features have shown thatcombining different descriptors can outperform the sensibilityin image classification.[13] A.X. Falcão, J. Stolfi, and R.A. Lotufo, “The image foresting transformtheory, algorithms, and applications,” IEEE Transactions on PatternAnalysis and Machine Intelligence, vol. 26, no. 1, pp. 19–29, 2004.[14] A. Mansano, J. A. Matsuoka, L. C S Afonso, J. P. Papa, F. Faria, andR. da S Torres, “Improving image classification through descriptorcombination,” in Graphics, Patterns and Images (SIBGRAPI), 201225th SIBGRAPI Conference on, 2012, pp. 324–329.[15] F. A. Faria, J. A. dos Santos, A. Rocha, and R. S. Torres, “Automaticclassifier fusion for produce recognition.,” in SIBGRAPI. 2012, pp. 252–259, IEEE Computer Society.[16] O.A.B. Penatti, E. Valle, and R.S. Torres, “Comparative study of globalcolor and texture descriptors for web image retrieval,” J. of VisualCommunication and Image Representation, 2011.VIII. ACKNOWLEDGMENTSThe authors thank FAPESP (Procs. #2011/11777-0,#2012/09809-4 and #2009/16206-1) and CNPq grant#303182/2011-3.REFERENCES[1] P. Gehler and S. Nowozin, “On feature combination for multiclass objectclassification,” in Proceedings of the 12th International Conference onComputer Vision, 2009, pp. 221–228.[2] J. Hou, B.-P. Zhang, N.-M. Qi, and Y. Yang, “Evaluating feature combinationin object classification,” in Proceedings of the 7th InternationalConference on Visual Computing, Las Vegas, NV, USA, 2011, pp. 597–606, Springer-Verlag.[3] D. Okanohara and J. Tsujii, “Learning combination features with L1regularization,” in Proceedings of Human Language Technologies: The2009 Annual Conference of the North American Chapter of the ACL,Stroudsburg, PA, USA, 2009, pp. 97–100.[4] X. Liu, L. Zhang, M. Li, H. Zhang, and D.Wang, “Boosting image classificationwith LDA-based feature combination for digital photographmanagement,” Pattern Recognition, vol. 38, no. 6, pp. 887–901, June2005.[5] F. Yan and X. Yanming, “Image classification based on multi-featurecombination and PCA-RBaggSVM,” in Proceedings of the IEEEInternational Conference on Progress in Informatics and Computing,2010, vol. 2, pp. 888–891.[6] F.A. Faria, J.P. Papa, R.S. Torres, and A.X. Falcão, “Multimodalpattern recognition through particle swarm optimization,” in Proceedingsof the 17th International Conference on Systems, Signals and ImageProcessing, Rio de Janeiro, Brazil, 2010, pp. 1–4.[7] Xin-She Yang, “Firefly algorithm, lévy flights and global optimization,”in SGAI Conf., 2009, pp. 209–218.[8] Z. Geem, “Novel derivative of harmony search algorithm for discretedesign variables,” Applied Mathematics and Computation, vol. 199, no.1, pp. 223–230, May 2008.[9] J. Kennedy and R.C. Eberhart, Swarm Intelligence, M. Kaufman, 2001.[10] J. P. Papa, A. X. Falcão, and C. T. N. Suzuki, “Supervised patternclassification based on optimum-path forest,” International Journal ofImaging Systems and Technology, vol. 19, no. 2, pp. 120–131, 2009.[11] J. P. Papa, A. X. Falcão, V. H. C. Albuquerque, and J. M. R. S.Tavares, “Efficient supervised optimum-path forest classification forlarge datasets,” Pattern Recognition, vol. 45, no. 1, pp. 512–520, 2012.[12] E. W. Dijkstra, “A note on two problems in connexion with graphs,”Numerische Mathematik, vol. 1, pp. 269–271, 1959.