. Tableau-4, La moyenne de l'erreur en %, procédure HS, pour la base divisée en deux pour la procédure « Un cotre Tous » des SVM. Les valeurs entre () affichent la variance de l'erreur Base Urbaine Intérieur Ouverte Fermée Cs-SVM 10

. Cs-lssvm, 14) 16, p.33

. Tableau-4, 13 : Erreur de classement en %, procédure HS, pour la base totale 1NN

L. Modèles and C. , Cs-SVM et CS-LSSVM surpassent le classifieur KNN. La différence est significative (test t de Student, p > 0,001) Une observation mal classée apporte 0

A. R. Kountchev and . Manolova, Fast algorithm for color space K-L image transform. Intern, Scientific Conf. on Information, Communication and Energy Systems and Technologies (ICEST " 05), pp.322-325, 2005.

A. R. Kountchev and . Manolova, Determination of Distances between Images Using Consecutive Approximations in the Linear Transforms Domain, Proc. of National Conf. with Foreign Participation (TELECOM " 05), 2005.

A. R. Kountchev and . Manolova, General Approach for Fast Image Retrieval from Image Database using Consecutive Iterations in the Transform Domain, Proceedings of the Technical University -Sofia, pp.167-176, 2005.

A. Manolova and . Guerin-dugue, Une nouvelle metrique pour l " analyse discriminante sur donnees de dissimilitude, 39ièmes Journées de Statistique, pp.11-15, 2007.
URL : https://hal.archives-ouvertes.fr/hal-00370450

A. Manolova, G. Celeux, and A. Guerin-dugue, Classification of dissimilarity data with a new flexible Mahalanobis like metric, PAA Special Issue on Non-parametric Distancebased Classification Techniques and their Applications, pp.337-351, 2008.

A. Manolova, R. Kountchev, and I. Aleksieva, Improved system for content based image retrieval based on pyramid decomposition in the spectrum domain, Intern, Scientific Conf. on Information, Communication and Energy Systems and Technologies (ICEST " 09), pp.25-27, 2009.

A. Manolova and A. Guerin-dugue, Dissimilarity based metric for data classification based on Support Vector Classifiers, XVI Rencontres de la Societe Francophone de Classification, pp.2-4, 2009.

A. Manolova and R. Kountchev, Face Identification with Modified K-NN classifier based on linear and angular distance, Intern, Scientific Conf. on Information, Communication and Energy Systems and Technologies (ICEST " 10), pp.23-26, 2010.

A. Manolova, Analyse discriminante sur tableaux de dissimilitudes : Application à la catégorisation d'images, 2004.

R. Bibliographiques, ]. S. Abe-]-e, . H. Adelson-]-e, E. P. Adelson, W. T. Simoncelli et al., Support vector machines for pattern classification Layered Representations for Image Coding, MIT Media Laboratory Vision and Modeling Technical Report Pyramids and Multiscale Representations, Representations of Vision Multiresolution image analysis: Content Based Image retrieval, pp.3-16, 1991.

]. E. Alpaydin-]-h, G. Bensmail, and . Celeux, Introduction to machine learningRegularized Gaussian discriminant analysis through eigenvalue decomposition, Journal of the American Statistical Association, vol.91, pp.1743-1791, 1996.

C. Biernacki, G. Celeux, G. Govaert, G. , and F. Langrognet, Le logiciel MIXMOD d'analyse de mélange pour la classification et l'analyse discriminante, La Revue de Modulad, vol.35, pp.25-44, 2006.

A. , D. Bimbo, ]. I. Borg, and P. Groenenboser, Visual Information Retrieval Pattern recognition and machine learning Modern multidimensional scaling: theory and applications A Training Algorithm for Optimal Margin Classifiers Towards bridging the gap between statistical and structural pattern recognition: Two new concepts in graph matching A tutorial on Support vector machines for pattern recognition, Data mining and knowledge discovery, International Conference on Advances in Pattern Recognition Image indexing and retrieval by content. Information Services and Use, pp.144-152, 1992.

]. G. Celeux and G. Govaert, Gaussian parsimonious clustering models, Celeux 1990] G. Celeux. Analyse Discriminante sur variables continues. Collection Didactique, pp.781-793, 1990.
DOI : 10.1016/0031-3203(94)00125-6

URL : https://hal.archives-ouvertes.fr/inria-00074643

]. G. Celeux and J. C. Turlot, Estimation de la qualité d'une règle discriminante, La Revue de Modulad, vol.4, pp.37-46, 1989.

]. T. Cox, M. A. Cox, N. Cristianini, and J. Shawe-taylor, Multidimensional scaling An Introduction to Support Vector Machines and Other Kernel-based Learning Methods Classification and dissimilarity analysis, Lecture notes in statisticsDerrode et al. 1999] S. Derrode, M. Daoudi, F. Ghorbel. Invariant Content-Based Image Retrieval Using a Complete Set of Fourier-Mellin Descriptors, Proceedings of the 1999 IEEE International Conference on Multimedia Computing and Systems, 1994.

S. Derrode, F. Ghorbel, ]. S. Derrode, F. Ghorbel, L. Györfi et al., Shape analysis and symmetry detection in graylevel objects using the analytical Fourier-Mellin representation Robust and efficient Fourier-Mellin transform approximations for gray-level image reconstruction and complete invariant description A probabilistic theory of pattern recognition, Deza, E. Deza. Encyclopedia of Distances, pp.25-39, 1996.

K. Duan and S. Keerthi, Which Is the Best Multiclass SVM Method? An Empirical Study, Proceedings of the Sixth International Workshop on Multiple Classifier Systems, 2005.
DOI : 10.1007/11494683_28

]. M. Dubuisson, A. K. Jain, and ]. Duchduda, Modified hausdorff distance for object matching Similarity-based methods: a general framework for classification, approximation and association, 12 th International Conference on Pattern Recognition, pp.566-568937, 1994.

. Duff, Direct methods for sparce matrices, Clarendon press Duin and E. P?kalska, Datasets and tools for dissimilarity analysis in pattern recognition, pp.1-174, 1986.

. Duin, On Euclidean corrections for non-Euclidean dissimilarities, Joint IAPR International Workshops on Statistical and Structural Pattern Recognition The dissimilarity representation, a basis for domain based pattern recognition?, invited talk (refereed) Representations in Pattern Recognition, IAPR Workshop Featureless pattern classification Representation is representation of similarities Skin detection ? a short tutorial, Fukunaga, Introduction to statistical pattern recognition Principles of Visual Information Retrieval, M. Lew, Ed., ch. Color in image search engines, pp.551-561, 1995.

L. Goldfarb, O. Golubitsky, D. Korkin, U. N. Goldberger, S. Gordon et al., An efficient image similarity measure based on approximations of KL-divergence between two gaussian mixtures Properties of Euclidean and non-Euclidean distance matrices, Linear algebra and its applications Fast Contour Matching Using Approximate Earth Mover " s Distance Discriminant analysis on dissimilarity data: a new fast Gaussian-like algorithm, Proceedings Ninth IEEE International Conference on Computer Vision Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Haasdonk, Transformation Knowledge in Pattern Analysis with Kernel Methods, pp.487-49381, 1986.

H. Haasdonk, . Burkhardt-]-b, D. Haasdonk, . Keysers-]-b, E. Haasdonk et al., Classification with Invariant Distance Substitution Kernels Data Analysis Tangent Distance Kernels for Support Vector Machines Classification with Kernel Mahalanobis Distance Classifiers, German Classification Society Annual Conference Indefinite Kernel Fisher Discriminant Classification using non-standard metrics The elements of statistical learning: Data mining, inference and prediction Learning kernel classifiers: Theory and algorithms Classification with Non- Metric Distances: Image Retrieval and Class Representation Fast Sparse Approximation for Least Squares Support Vector Machine, Feature Space Interpretation of SVMs with Indefinite Kernels. IEEE Transactions on Pattern Analysis and Machine Intelligence Bahlmann. Learning with Distance Substitution Kernels. Pattern Recognition -Proc. of the 26th DAGM Symposium Proc. of 31st. GfKl Conference Machine Learning, and Applications Proc. of the 16th Int. Conf. on Pattern RecognitionHan&Kamber 2006] J. Han, M. Kamber. Data Mining: Concepts and TechniquesHuttenlocher et al. 1993] D. Huttenlocher, G. Klanderman and W. Rucklidge. Comparing Images Using the Hausdorff Distance Making large-Scale SVM Learning Practical. Advances in Kernel Methods -Support Vector LearningKeshavarzi et al. 2009] M. Keshavarzi, M. A. Deghan, M. Mashinchi. Classification based on similarity and dissimilarity through equivalence classes, Appl. and Comput. Math., V.8, pp.482-492, 1993.

]. A. Koening, Analyse discriminante sur tableaux de dissimilarités, 2000.

. Kountchev, Inverse Pyramidal Decomposition with multiple DCT, Signal processing, Kountchev et al. 2005] Multi-layer Image Transmission with Inverse Pyramidal Decomposition. Computational Intelligence for Modelling and Prediction 2005, pp.179-196, 2002.

. Kountchev, Invariant object representation with modified Mellin-Fourier transform, J. B. Kruskal, M. Wish. Multidimensional scaling, pp.232-236, 1978.

]. D. Larose and Z. Mackay, Data mining: methods and models A Probabilistic Model for the Multidimensional Scaling of Proximity and Preference Data, Marketing Science, Machine learning, Neural and statistical classification Milanese, M. Cherbuliez. A Rotation, Translation, and Scale-Invariant Approach to Content-Based Image Retrieval, pp.186-196, 1986.

. Moreno, A Kullback?Leibler divergence based kernel for SVM classification in multimedia applications Information retrieval for music and motion, Neural information processing system, 2003.

]. R. Niels-]-m, A. S. Nixon, and . Aguadoosuna, Dynamic Time Warping: An Intuitive Way of Handwriting Recognition?, Master's thesis Feature extraction and image processing, pp.130-136, 1997.

]. E. P?kalska-]-e, B. P?kalska, . Haasdonk-]-e, R. P. P?kalska, and . Duin, The Dissimilarity representations in pattern recognition. Concepts, theory and applications, ASCI Dissertation Series no. 109 Kernel Discriminant Analysis for Positive Definite and Indefinite Kernels Beyond traditional kernels: classification in two dissimilarity-based representation spaces The dissimilarity representation for pattern recognition, IEEE Transactions on Pattern Analysis and Machine Intelligence IEEE Transactions on Systems, Man and Cybernetics--Part C Foundations and Applications, vol.31, issue.38 6, pp.729-744, 1017.

. Pekalska, Prototype Selection for Dissimilaritybased Classifiers Dissimilarity representations allow for building good classifiers A Generalized Kernel Approach to Dissimilarity-based Classification Automatic pattern recognition by similarity representations Dissimilarity-based classification with vectorial representations, oral presentation, International Conference on Pattern RecognitionPekalska et al. 2005] E.P?kalska, A.Harol, C.Lai and R.P.W. Duin. Pairwise selection of features and prototypes, oral presentation, International Conference on Computer Recognition SystemsPetres 2008] R.A. Peters II. On the Computation of the Discrete Log-Polar Transform, pp.189-208, 2001.

. Piro, Similarity between Euclidean and cosine angle distance for nearest neighbor queries The earth mover " s distance as a metric for image retrieval On using prototype reduction schemes to optimize dissimilarity-based classification In Defense of One-Vs-All Classification A Survey of Methods for Color Image Indexing and Retrieval in Image Databases, ch. Color imaging science: Exploiting digital media New support vector algorithms Computational studies of the spatial architecture of primate visual cortex: Columns, maps, and protomaps, CBMI Proc. of the 2004 ACM symposium on applied computing Tests de normalité: Techniques empiriques et tests statistiques Arificial neural networks The kernel trick for distances. In: Neural information processing system Primary Visual Cortex in Primates, pp.1232-1237, 1994.

. Simard, Transformation invariance in pattern recognition ? tangent distance and tangent propagation, Neural Networks: Tricks of the Trade, pp.239-266, 1996.

. Smeulders, Content-based image retrieval at the end of the early years, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.22, issue.12, 2000.
DOI : 10.1109/34.895972

M. Anal and . Intelligencesuykens, Least squares support vector machine classifiers: a large scale algorithm, Proc. of the European Conference on Circuit Theory and Design (ECCTD'99), pp.1349-1380, 1999.

J. A. Suykens and J. Vandewalle, Recurrent least squares support vector machines, IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications, vol.47, issue.7, pp.1109-1114, 2000.
DOI : 10.1109/81.855471

J. A. Suykens, J. Vandewalle, and B. De-moor, Optimal control by least squares support vector machines, Neural Networks, vol.14, issue.1, pp.23-35, 2001.
DOI : 10.1016/S0893-6080(00)00077-0

URL : http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.20.9496

. Suykens, Least Squares Support Vector Machines, 2002.
DOI : 10.1142/5089

J. A. Suykens, G. Horvath, S. Basu, C. Micchelli, J. S. Vandewalle et al., Advances in Learning Theory : Methods, Models and Applications The Netherlands), (ISBN 1-58603-341-7), NATO-ASI Series III : Computer and Systems SciencesTypke et al. 2003] R. Typke, P. Giannopoulos, R.C. Veltkamp, F. Wiering and R. van Oostrum. Using transportation distances for measuring melodic similarity, 1973.

]. R. Typke, Music Retrieval based on Melodic Similarity, Proc. ISMIR 2003, pp.107-114, 2003.

B. Dedene, . De, J. Moor, and . Vandewalle, Benchmarking least squares support vector machine classifiers [Vapnik 1995] V. Vapnik. The Nature of Statisticsal Learning, Machine LearningVezhnevets et al. 2003] V. Vezhnevets, V. Sazonov, A.Andreeva. A survey on pixel-based skin color detection techniques, Proc. Graphicon-2003, pp.5-32, 1995.

. Wang, On Euclidean distance for images, Pattern Analysis and Machine Intelligence, IEEE, issue.8, pp.27-1334, 2005.

Q. Wang, Y. Li, K. Zhang, A. R. Xian-ze-xiong, . Webb-]-g et al., Inverse log-polar transformation algorithm based on sub-pixel interpolation, Optoelectronics Letters Statistical pattern recognition The effect of class distribution on classifier learning : on empirical study Face recognition based on polar frequency features, ACM Transactions on Applied Perception, vol.2, issue.31, pp.62-82, 2001.

]. A. Zinke and . Mayer, Iterative Multi Scale Dynamic Time Warping, Zoutendijk. Nonlinear Programming Computational Methods. In: Integer and Nonlinear Programming, pp.37-86, 1970.