OPEN ACCESS
Content-based image retrieval in large image data sets is a tiresome task considering the high rate of heterogeneity even within the same class as well as the high dimensionality of the descriptors space. For that, we propose to guide the research by low level descriptors of reduced size based on sub-bands of wavelet relating to the most significant regions in each image after a fuzzy segmentation step. Moreover, we propose a negative relevance feedback technique based on region weights. Experiments and comparative study with other similar approaches prove the robustness of the proposed approach in terms of semantic contribution thanks to the use of the wavelet sub-band and the negative relevance feedback.
RÉSUMÉ
La recherche des images par le contenu dans des grandes bases généralistes d’images est une tâche fastidieuse vu le taux d’hétérogénéité assez élevé même au sein d’une seule classe de la base ainsi que la grande dimension de l’espace des descripteurs relatifs aux images. Pour cela, nous proposons de guider la recherche par des descripteurs de bas-niveau de taille réduite à base de sous-bandes d’ondelettes relatives aux régions les plus significatives dans chaque image après une phase de segmentation floue. De plus, nous proposons une technique de bouclage de pertinence négatif sur les poids relatifs aux régions. Les expérimentations et l’étude comparative avec des approches similaires prouvent la robustesse de l’approche proposée en termes d’apport sémantique offert par l’utilisation des sous-bandes d’ondelettes ainsi que par le bouclage de pertinence négatif.
region-based image retrieval, negative relevance feedback, wavelet transformation, HH wavelet sub-band, region weights
MOTS-CLÉS
recherche des images par le contenu, bouclage de pertinence négatif, transformation d’ondelettes, sous-bande HH, pondération des régions
Barhoumi W., Gallas A., Zagrouba E. (2009). Effective region-based relevance feedback for interactive content-based image retrieval. New directions in intelligent interactive multimedia systems and services, vol. 226, p. 177-187.
Borghesani D., Grana C., Cucchiara R. (2012). Towards Artistic Collections Navigation Tools Based on Relevance Feedback. Multimedia for cultural heritage, vol. 247, p. 143-153.
Bulò S.R., Rabbi M., Pelillo M. (2011). Content-based image retrieval with relevance feedback using random walks. Pattern recognition, vol. 44, n° 9, p. 2109-2122.
Chen Z. (2010). Semantic Research on Content-Based Image Retrieval. International Conference on Multimedia Technology. Ningbo, 29-31 October 2010. p. 1-4.
Cox I.J., Minka T.P., Papathomas T.V., Yianilos P.N. (2000). The Bayesian image retrieval system, PicHunter: Theory, implementation, and psychophysical experiments. IEEE transactions on image processing, vol. 9, n° l, p. 20-37.
Deselaers T., Keyserse D., Ney H. (2004). Features for image retrieval: a quantitative comparison. Notes in computer science, vol. 3175, p. 228-236.
Estevez D. (2001). Dessin d’architecture et infographie, L’évolution contemporaine des pratiques graphiques. Paris, France, CNRS Editions.
Francisco da Silva S., Xavier Ribeiro M., Batista Neto J., Traina-Jr C., Traina A. (2011). Improving the ranking quality of medical image retrieval using a genetic feature selection method. Decision support systems, vol. 50, p. 1-11.
Gallas A., Barhoumi W., Zagrouba E. (2010). Une approche rapide basée sur les régions floues pour une évaluation sémantique de la similarité d’images. WIA 2010 Workshop en informatique et applications. Tozeur, Tunisie, 12-14 février.
Hiremath P.S., Shivashankar S., Pujari J.D. (2006). Wavelet based features for color texture classification with application to CBIR. Journal of computer science and network security, vol. 6, n° 9, p. 124-133.
Hsu C.T., Li C.Y. (2005). Relevance feedback using generalized Bayesian framework with region-based optimization learning. IEEE transactions on image processing, vol. 14, n° 1, p. 1617-1631.
Hua K.A., Yu N., Liu D. (2006). Query decomposition: a multiple neighborhood approach to relevance feedback processing in content-based image retrieval. International conference on data engineering. Atlanta, 3-8 Avril.
Huang W., Gao Y., Chan K.L. (2008). A review of region-based image retrieval. Sign process system, vol. 59, p. 143-161.
Jing F., Li M., Zhang L., Zhang H.J., Zhang B. (2003). Learning in region-based image retrieval. Conference on image and video retrieval. Urbana-Champaign, 24-25 Juillet,p. 206-215.
Kao O. (2008). On parallel image retrieval with dynamically extracted features. Parallel computing, vol. 34, p. 700-709.
Kherfi M.L., Ziou D. (2006). Relevance feedback for CBIR: a new approach based on probabilistic feature weighting with positive and negative examples. IEEE transactions on image processing, vol. 15, n° 4, p.1017-1030.
Kim D.H., Yu S.H. (2008).A new region filtering and region weighting approach to relevance feedback in content-based image retrieval. The journal of systems and software, vol. 81,p. 1525-1538.
Kim D.H., Song J.W., Lee J.H. (2007). A hybrid region weighting approach for relevance feedback in region-based image search on the web. Lecture notes in computer science, vol. 4362, p. 705-715.
Lamard M., Cazuguel G., Quellec G., Bekri L., Roux C., Cochener B. (2007). Content based image retrieval based on wavelet transform coefficients distribution. The annual international conference of the IEEE engineering in medicine and biology society. Lyon, France, 23-26 Août.
Li J., Wang J.Z. (2008). Real-time computerized annotation of pictures. IEEE transactions on pattern analysis and machine intelligence, vol. 30, n° 6, p. 985-1002.
Liua R., Wanga Y., Babab T., Masumotob D., Nagatab S. (2008). SVM-based active feedback in image retrieval using clustering and unlabeled data. Pattern recognition,vol. 41, p. 2645-2655.
Paredes R., Deselaers T., Vidal E. (2008). A probabilistic model for user relevance feedback on image retrieval. The joint workshop on machine learning and multimodal interaction. Utrecht, 8-10 Septembre, p. 260-271.
Philipp-Foliguet S., Vieira M.B., Lekkat M., (2005). Image retrieval by fuzzy regions sets matching. Information interaction intelligence, vol. 5, n° 2, p. 9-39.
Piatrik T., Izquierdo E. (2011). Multi-feature fusion in image clustering using ant-inspired methods. Third orld congress on nature and biologically inspired computing, Salamanca, 19-21 October, p. 377-382.
Su Z., Zhang H.J., Li S., Ma S. (2003). Relevance feedback in content-based image retrieval: Bayesian framework feature sub-spaces and progressive learning. IEEE transactions on image processing, vol. 12, n° 8, p. 924-937.
Tian Q., Sebe N., Lew M.S., Loupias E., Huang T.S. (2001). Content-based image retrieval using wavelet-based salient points. Storage and retrieval for media databases. San Jose, 24 Janvier, p. 425-436.
Vannoorenberghe P. (2004). Bouclage de pertinence par arbres de décision crédibilistes en indexation d’images par le contenu. Compression et représentation des signaux audiovisuels, Lille, 25- 26 mai.
Wang J.Z., Li J., Wiederhold G. (2001). SIMPLIcity: Semantics-sensitive integrated matching for picture libraries. IEEE Transactions on pattern analysis and machine intelligence, vol. 23, n° 9, p. 947-963.
Xu X., Lee D., Antani S.K., Long L.R., Archibald J.K. (2009). Using relevance feedback with short-term memory for content-based spine X-ray image retrieval. Neurocomputing, vol. 71, n° 10, p. 2259-2269.
Zhang C., Ouyang D., Ning J. (2010). An artificial bee colony approach for clustering. Expert systems with applications, vol. 37, p. 4761-4767.
Zhang Z., Toda H., Kawabata H. (2002). A new complex wavelet: RI-Spline wavelet and its application to signal processing. The annual conference of society of instrument and control engineers. Osaka, 5-7 Août, p. 2496-2501.