Jimmy Francky Randrianasoa

New algorithms for multivalued component trees

By Nicolas Passat, Romain Perrin, Jimmy Francky Randrianasoa, Camille Kurtz, Benoît Naegel

2024-09-30

In Proceedings of the 27th international conference on pattern recognition

Abstract

Tree-based structures can model images—and more generally valued graphs—for processing and analysis purpose. In this framework, the component tree was natively designed for grey-level images—and more generally totally or- dered valued graphs. Ten years ago, the notion of a multivalued component tree was introduced to relax this grey-level / total order constraint. In this algorith- mic paper, we provide new tools to handle multivalued component trees. Our contributions are twofold: (1) we propose a new algorithm for the construction of the multivalued component tree; (2) we propose two strategies for building hierarchical orders on value sets, required to further build the multivalued com- ponent trees of images / graphs relying on such value sets. Codes available at: https://github.com/bnaegel/multivalued_component_tree.

Continue reading

New algorithms for multivalued component trees

By Nicolas Passat, Romain Perrin, Jimmy Francky Randrianasoa, Camille Kurtz, Benoît Naegel

2024-06-30

In

Abstract

Les structures arborescentes peuvent modéliser les images—et plus généralement les graphes valués—à des fins de traitement et d’analyse. Dans ce cadre, l’arbre des coupes a été nativement proposé pour les images en niveaux de gris et, plus généralement, pour les graphes valués totalement ordonnés. Il y a dix ans, la notion d’arbre des coupes multivalués a été introduite pour assouplir cette contrainte de niveaux de gris / ordre total. Dans cet article algorithmique, nous fournissons de nouveaux outils pour traiter les arbres des coupes multivalués. Nos contributions sont doubles : (1) nous proposons un nouvel algorithme pour la construction de l’arbre des coupes multivalués ; (2) nous proposons deux stratégies pour construire des ordres hiérarchiques sur des ensembles de valeurs, nécessaires pour construire les arbres des coupes multivalués d’images / de graphes reposant sur de tels ensembles de valeurs.

Continue reading

AGAT: Building and evaluating binary partition trees for image segmentation

By Jimmy Francky Randrianasoa, Camille Kurtz, Éric Desjardin, Nicolas Passat

2021-12-17

In SoftwareX

Abstract

AGAT is a Java library dedicated to the construction, handling and evaluation of binary partition trees, a hierarchical data structure providing multiscale partitioning of images and, more generally, of valued graphs. On the one hand, this library offers functionalities to build binary partition trees in the usual way, but also to define multifeature trees, a novel and richer paradigm of binary partition trees built from multiple images and/or several criteria. On the other hand, it also allows one to manipulate the binary partition trees, for instance by defining/computing tree cuts that can be interpreted in particular as segmentations when dealing with image modeling. In addition, some evaluation tools are also provided, which allow one to evaluate the quality of different binary partition trees for such segmentation tasks. AGAT can be easily handled by various kinds of users (students, researchers, practitioners). It can be used as is for experimental purposes, but can also form a basis for the development of new methods and paradigms for construction, use and intensive evaluation of binary partition trees. Beyond the usual imaging applications, its underlying structure also allows for more general developments in graph-based analysis, leading to a wide range of potential applications in computer vision, image/data analysis and machine learning.

Continue reading

Search

Tags