This paper presents a novel approach to combine shape descriptors. Each approach is applied on several clusters of objects. For each cluster and for any descriptor a map is as-sociated directly from the confusion matrix. Such a method allows to determine automatically the better weight asso-ciated to the descriptor for the object under consideration. At last, we show that the additive combination of such mea-sures allows to improve the classification.
Jean-Pierre Salmon, Laurent Wendling, Salvatore Tabbone. Automatical Definition of Measures from the Combination of Shape Descriptors. 8th International Conference on Document Analysis and Recognition - ICDAR'2005, Aug 2005, Seoul/Korea, pp.986-990. ⟨inria-00000513⟩ (lien externe)
Citations
APA
Salmon, J.-P., Wendling, L., & Tabbone, S. (2005). Automatical Definition of Measures from the Combination of Shape Descriptors. https://inria.hal.science/inria-00000513v1
MLA
Salmon, Jean-Pierre, et al. Automatical Definition of Measures from the Combination of Shape Descriptors. Aug. 2005, https://inria.hal.science/inria-00000513v1.
Chicago
Salmon, Jean-Pierre, Laurent Wendling, and Salvatore Tabbone. 2005. Automatical Definition of Measures from the Combination of Shape Descriptors. https://inria.hal.science/inria-00000513v1.
Harvard
Salmon, J.-P., Wendling, L. and Tabbone, S. (2005) “Automatical Definition of Measures from the Combination of Shape Descriptors.” Available at: https://inria.hal.science/inria-00000513v1.
ISO 690
SALMON, Jean-Pierre, WENDLING, Laurent and TABBONE, Salvatore, 2005. Automatical Definition of Measures from the Combination of Shape Descriptors [en ligne]. August 2005. Disponible à l'adresse : https://inria.hal.science/inria-00000513v1