Learning prototype-based classification rules in a boosting framework: application to real-world and medical image categorization

Paolo Piro (1)

1 Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe IMAGES-CREATIVE

In this PhD thesis we address some major topics related to indexing and classification of images. In particular, we investigate the most relevant functional blocks of an image retrieval/categorization system, and propose original solutions to address some of their specific issues.

Résumé en français non disponible

Mots clés: images; indexation; classification; Sparse Multiscale Patches (SMP); knearest neighbor (k-NN); Universal Nearest Neighbors (UNN); MLNN algorithm; medical image classification; radiographic images

Domaines: Interface homme-machine [cs.HC]

Paolo Piro. Learning prototype-based classification rules in a boosting framework: application to real-world and medical image categorization. Human-Computer Interaction [cs.HC]. Université Nice Sophia Antipolis, 2010. English. ⟨NNT : ⟩. ⟨tel-00590403⟩