May 2015 - Annonce

New publication: How to Make nD Images Well-composed without Interpolation (ICIP'15)
by nboutry 14 May '15

14 May '15

I am happy to announce that the following paper has been accepted at the 22th IEEE International Conference on Image Processing (ICIP'2015), to be held on September 27-30 in Québec City, Canada. How to make nD images well-composed without interpolation Nicolas Boutry¹², Thierry Géraud¹, Laurent Najman² ¹ EPITA Research and Development Laboratory (LRDE) ² Université Paris-Est, LIGM, Équipe A3SI, ESIEE Paris https://www.lrde.epita.fr/wiki/Publications/boutry.15.icip Abstract: Latecki et al. have introduced the notion of well-composed images, i.e., a class of images free from the connectivities paradox of discrete topology. Unfortunately natural and synthetic images are not a priori well-composed, usually leading to topological issues. Making any nD image well-composed is interesting because, after- wards, the classical connectivities of components are equivalent, the component boundaries satisfy the Jordan separation theorem, and so on. In this paper, we propose an algorithm able to make nD images well-composed without any interpolation. We illustrate on text detection the benefits of having strong topological properties.

1 0

Séminaire MeFoSyLoMa du vendredi 22 mai 2015, au LRDE
by Alexandre Duret-Lutz 12 May '15

12 May '15

======================================================================== Séminaire MeFoSyLoMa http://www.mefosyloma.fr/ Méthodes Formelles pour les Systèmes Logiciels et Matériels vendredi 22 mai 2015, 14h-17h Adresse: LRDE (EPITA), Salle Alpha Bâtiment X, 2e étage 18 rue Pasteur, 94270 Le Kremlin-Bicêtre Métro Porte d'Italie Plans d'accès : https://www.google.com/maps?q=48.8152808,2.3623765 https://www.lrde.epita.fr/~adl/dl/acces-lrde.pdf S'il y en a parmi vous qui comptent venir en voiture, le parking du centre commercial Okabé est gratuit pendant 3h. http://www.okabe.com/okabe/fr/okabe-parking ======================================================================== Le séminaire MeFoSyLoMa est animé conjointement par les laboratoires Cedric (Cnam), IBISC (Univ. Evry), LACL (Univ. Paris 12), LIP6 (UPMC), LIPN (Univ. Paris 13), LRDE (Epita), LSV (École Normale Supérieure de Cachan) et LTCI (TELECOM ParisTech). Son objet est de permettre la confrontation de différentes approches ou points de vue sur l'utilisation des méthodes formelles dans les domaines du génie logiciel, de la conception de circuit, des systèmes répartis, des systèmes temps-réel ou encore des systèmes d'information. Il s'organise autour de réunions bimestrielles où sont exposés des travaux de recherche récents sur ce thème. ======================================================================== Programme 14h00-15h00: Ekkart Kindler, Denmark Technical University, "Coordinating Interactions: The Event Coordination Notation" Abstract: The purpose of a domain model is to concisely capture the concepts of an application's domain, and their relation among each other. Even though the main purpose of domain models is not implementing the application, major parts of an application can be generated from the application's domain models fully automatically with today's technologies. The focus of most of today's code generation technologies, however, is on the structural aspects of the domain; the domain's behaviour is often not modelled at all, implemented manually based on some informal models, or the behaviour is modelled on a much much more technical level. The Event Coordination Notation (ECNO) allows modelling the behaviour of an application on a high level of abstraction that is closer to the application's domain than to the software realizing it. Still, these models contain all necessary details for actually executing the models and for generating code from them. In this talk, the limitations of today's modelling notations for behaviour are briefly discussed. Then, the main idea, philosophy, and concepts of ECNO and its notation are discussed -- mostly by looking at some examples. The ECNO is now fully supported by a tool which allows to generate code from ECNO models. 15h00-16h00: Ryszard Janicki, McMaster University, "Modeling Concurrency With Interval Traces" Abstract: Interval order structures are triples (X,≺,⊏) where X is a set of event occurences and ≺, ⊏ are abstractions of 'earlier than' and 'not later than' relationships. Interval order structures are useful tools to model abstract concurrent histories, i.e. sets of equivalent system runs, when system runs are modeled with interval orders and we want to express not only standard causality but also 'not later than'. It turns out that interval order structures can be modeled by partially commutative monoids, called interval traces, that are some special case of general Mazurkiewicz traces. This new model will then be used to provide a full semantics of Petri nets with inhibitor arcs. 16h00-16h30: pause café 16h30-17h00: vie du groupe ========================================================================

1 0

Paper accepted at COP'2015 (ECOOP)
by Didier Verna 11 May '15

11 May '15

Hello, we're happy to announce that our paper entitled "Context-Oriented Image Processing" has been accepted at ECOOP's Context-Oriented Programming Workshop (COP'2015), to be held in Prague this July. The abstract is given below. Genericity aims at providing a very high level of abstraction in order, for instance, to separate the general shape of an algorithm from specific implementation details. Reaching a high level of genericity through regular object-oriented techniques has two major drawbacks, however: code cluttering (e.g. class / method proliferation) and performance degradation (e.g. dynamic dispatch). In this paper, we explore a potential use for the Context-Oriented programming paradigm in order to maintain a high level of genericity in an experimental image processing library, without sacrificing either the performance or the original object-oriented design of the application. -- My new Jazz CD entitled "Roots and Leaves" is out! Check it out: http://didierverna.com/records/roots-and-leaves.php Lisp, Jazz, Aïkido: http://www.didierverna.info

1 0

New publication: How to Make nD Functions Digitally Well-Composed in a Self-Dual Way (ISMM'15)
by nboutry 05 May '15

05 May '15

I am happy to announce that the following paper has been accepted at the 12th International Symposium on Mathematical Morphology (ISMM'15), to be held on May 27-29 2015 in Reykjavik, Iceland. How to Make nD Functions Digitally Well-Composed in a Self-Dual Way Nicolas Boutry¹², Thierry Géraud¹, Laurent Najman² ¹ EPITA Research and Development Laboratory (LRDE) ² Université Paris-Est, LIGM, Équipe A3SI, ESIEE Paris https://www.lrde.epita.fr/wiki/Publications/boutry.15.ismm Abstract: Latecki et al. introduced the notion of 2D and 3D well-composed images, i.e., a class of images free from the ``connectivities pa- radox'' of digital topology. Unfortunately natural and synthetic images are not a priori well-composed. In this paper we extend the notion of ``digital well-composedness'' to nD sets, integer- valued functions (gray-level images), and interval-valued maps. We also prove that the digital well-composedness implies the equi- valence of connectivities of the level set components in nD. Con- trasting with a previous result stating that it is not possible to obtain a discrete nD self-dual digitally well-composed function with a local interpolation, we then propose and prove a self- dual discrete (non-local) interpolation method whose result is always a digitally well-composed function. This method is based on a sub-part of a quasi-linear algorithm that computes the morpholo- gical tree of shapes.

1 0

New publication: Efficient Computation of Attributes and Saliency Maps on Tree-Based Image Representations (ISMM'15)
by Edwin Carlinet 05 May '15

05 May '15

I am happy to announce that the following paper has been accepted at the 12th International Symposium on Mathematical Morphology (ISMM'15), to be held on May 27-29 2015 in Reykjavik, Iceland. I am happy to announce that the following paper has been accepted at the 12th International Symposium on Mathematical Morphology (ISMM'15), to be held on May 27-29 2015 in Reykjavik, Iceland. Efficient Computation of Attributes and Saliency Maps on Tree-Based Image Representations Yongchao Xu¹² and Edwin Carlinet¹² and Thierry Géraud¹ and Laurent Najman² ¹ EPITA Research and Development Laboratory (LRDE) ² Université Paris-Est, LIGM, Équipe A3SI, ESIEE Paris https://www.lrde.epita.fr/wiki/Publications/xu.15.ismm Abstract: Tree-based image representations are popular tools for many applications in mathematical morphology and image processing. Classically, one computes an attribute on each node of a tree and decides whether to preserve or remove some nodes upon the attribute function. This attribute function plays a key role for the good performance of tree-based applications. In this paper, we propose several algorithms to compute efficiently some attribute information. The first one is incremental computation of information on region, contour, and context. Then we show how to compute efficiently extremal information along the contour (e.g., minimal gradient's magnitude along the contour). Lastly, we depict computation of extinction-based saliency map using tree-based image representations. The computation complexity and the memory cost of these algorithms are analyzed. To the best of our knowledge, except information on region, none of the other algorithms is presented explicitly in any state-of-the-art paper. -- Edwin Carlinet

1 0

New publication: A Color Tree of Shapes with Illustrations on Filtering, Simplification, and Segmentation (ISMM'15)
by Edwin Carlinet 05 May '15

05 May '15

I am happy to announce that the following paper has been accepted at the 12th International Symposium on Mathematical Morphology (ISMM'15), to be held on May 27-29 2015 in Reykjavik, Iceland. A Color Tree of Shapes with Illustrations on Filtering, Simplification, and Segmentation Edwin Carlinet¹², Thierry Géraud¹ ¹ EPITA Research and Development Laboratory (LRDE) ² Université Paris-Est, LIGM, Équipe A3SI, ESIEE Paris https://www.lrde.epita.fr/wiki/Publications/carlinet.15.ismm Abstract: The Tree of Shapes is a morphological tree that provides a high-level, hierarchical, self-dual, and contrast invariant representation of images, suitable for many image processing tasks. When dealing with color images, one cannot use the Tree of Shapes because its definition is ill-formed on multivariate data. Common workarounds such as marginal processing, or imposing a total order on data are not satisfactory and yield many problems (color artifacts, loss of invariances, etc.) In this paper, we highlight the need for a self-dual and contrast invariant representation of color images and we provide a method that builds a single Tree of Shapes by merging the shapes computed marginally, while guarantying the most important properties of the ToS. This method does not try to impose an arbitrary total ordering on values but uses only the inclusion relationship between shapes. Eventually, we show the relevance of our method and our structure through some illustrations on filtering, image simplification, and interactive segmentation. -- Edwin Carlinet

1 0

New publication: Self-Duality and Digital Topology (ISMM'15)
by Thierry Géraud 05 May '15

05 May '15

I am happy to announce that the following paper has been accepted at the 12th International Symposium on Mathematical Morphology (ISMM'15), to be held on May 27-29 2015 in Reykjavik, Iceland. Self-Duality and Digital Topology: Links Between the Morphological Tree of Shapes and Well-Composed Gray-Level Images Thierry Géraud¹, Edwin Carlinet¹², Sébastien Crozet¹, ¹ EPITA Research and Development Laboratory (LRDE) ² Université Paris-Est, LIGM, Équipe A3SI, ESIEE Paris https://www.lrde.epita.fr/wiki/Publications/geraud.15.ismm Abstract: In digital topology, the use of a pair of connectivities is re- quired to avoid topological paradoxes. In mathematical morphology, self-dual operators and methods also rely on such a pair of connectivities. There are several major issues: self-duality is impure, the image graph structure depends on the image values, it impacts the way small objects and texture are processed, and so on. A sub-class of images defined on the cubical grid, well-composed images, has been proposed, where all connectivities are equivalent, thus avoiding many topological problems. In this paper we unveil the link existing between the notion of well-composed images and the morphological tree of shapes. We prove that a well-composed image has a well-defined tree of shapes. We also prove that the only self-dual well-composed interpolation of a 2D image is obtained by the median operator. What follows from our results is that we can have a purely self-dual representation of images, and consequently, purely self-dual operators.

1 0