Herman Akdag

Université Paris Viii, UFR MITSIC, Faculty Member

Followers

Following

Co-authors

Public Views

marie durand

Universite Paris-8, France

Saoussen Bel Hadj Kacem

Interests

Uploads

Papers by Herman Akdag

A Tool for Aggregation with Words

HAL (Le Centre pour la Communication Scientifique Directe), 2009

The need of computing with words has become an important topic in many areas dealing with vague i... more The need of computing with words has become an important topic in many areas dealing with vague information. The aim of this paper is to present different tools which support computing with words. Especially, we are concerned with the weighted aggregation of linguistic term sets, without using fuzzy concepts. We propose a new aggregation operator, referred to as the symbolic weighted median that computes the most representative element from an ordered collection of weighted linguistic terms. This operator aggregates the linguistic labels such that its result is expressed in terms of the initial linguistic term set though is modified by using dedicated tools called the generalized symbolic modifiers. One advantage of this proposal is that the expression domain does not change: we increase or decrease the granularity only where it becomes necessary. Additionally this new operator exhibits several interesting mathematical properties.

Download

A Tool for Aggregation with Words

HAL (Le Centre pour la Communication Scientifique Directe), 2009

Download

Image Retrieval using Fuzzy Representation of Colors

HAL (Le Centre pour la Communication Scientifique Directe), 2007

The problem addressed in this article is image indexing and retrieval according to the color. Ind... more The problem addressed in this article is image indexing and retrieval according to the color. Indeed we propose a classification based on the dominant color(s) of the images. The process consists in two steps: first, assigning a colorimetric profile to the image in HLS space (Hue, Lightness, Saturation) and then, handling the query for the retrieval. To achieve the first step, the definition of hue is done using a fuzzy representation that takes into account the non-uniformity of color distribution. Then, lightness and saturation are represented through linguistic qualifiers also defined in a fuzzy way. Finally, the profile is built through fuzzy functions representing the membership degree of the image to different classes. Thus, the query for image retrieval is a pair (hue, qualifier). The second step looks for a match between the query and the profiles. In order to improve the software and to make it more flexible, the user can redefine the fuzzy representation of Hue, Lightness and Saturation, according to his own perception.

Download

Approximate Reasoning in Supervised Classification Systems

Journal of Advanced Computational Intelligence and Intelligent Informatics, Jul 20, 2006

In knowledge-based systems, uncertainty in propositions can be represented by various degrees of ... more In knowledge-based systems, uncertainty in propositions can be represented by various degrees of belief encoded by numerical or symbolic values. The use of symbolic values is necessary in areas where the exact numerical values associated with a fact are unknown by experts. In this paper we present an expert system of supervised automatic classification based on a symbolic approach. This last is composed of two sub-systems. The first sub-system automatically generates the production rules using training set; the generated rules are accompanied by a symbolic degree of belief which characterizes their classes of memberships. The second is the inference system, which receives in entry the base of rules and the object to classify. Using classical reasoning (Modus Ponens), the inference system provides the membership class of this object with a certain symbolic degree of belief. Methods to evaluate the degree of belief are numerous but they are often tarnished with uncertainty. To appreciate the performances of our symbolic approach, tests are made on the Iris data basis.

Apprentissage supervisé et raisonnement approximatif, l'hypothèse des imperfections

HAL (Le Centre pour la Communication Scientifique Directe), 2001

Nous proposons une methode d'apprentissage a partir d'exemples qui se situe a la jonction... more Nous proposons une methode d'apprentissage a partir d'exemples qui se situe a la jonction des methodes statistiques et de celles basees sur des techniques d'intelligence artificielle. Notre modelisation se base sur la generation automatique de regles de classification et sur une utilisation originale du raisonnement approximatif. La methode d'apprentissage proposee, basee sur une recherche de correlations lineaires entre les composantes des vecteurs d'apprentissage est multi-attributs. L'incertitude des regles est geree aussi bien dans la phase d'apprentissage que dans celle de la reconnaissance. Un systeme baptise SUCRAGE a ete implemente et confronte a une application reelle dans le domaine du traitement d'images. Les resultats obtenus permettent de valider notre approche et nous autorisent a envisager d'autres domaines d'application. De plus, ces resultats confortent notre hypothese des imperfections: le raisonnement approximatif ou une recherche de correlations intraclasses peuvent sensiblement ameliorer les resultats.

A symbolic Approach for Colorimetric Alterations

HAL (Le Centre pour la Communication Scientifique Directe), Sep 1, 2001

To modify colours, many solutions are possible. Here, we present an approach that uses symbolic l... more To modify colours, many solutions are possible. Here, we present an approach that uses symbolic linguistic modifiers. Colours are modified thanks to two kinds of words: a modifier and a qualifier. The first one is translated into a symbolic linguistic modifier whereas the second one into one or many fuzzy subsets. In fact, we can say that we associate words with these two kinds of simple mathematical objects.

Supervised learning using modifiers: application in colorimetrics

Summary form only given. We present a method for a learning process that we apply to a colorimetr... more Summary form only given. We present a method for a learning process that we apply to a colorimetric application. In the application, we associate colours to linguistic expressions and these associations can be changed thanks to our learning process. The method consists of memorizing the meaning of the linguistic expressions according to a learning done thanks to the user. To

Extraction des couleurs dominantes d’une image

HAL (Le Centre pour la Communication Scientifique Directe), Mar 25, 2007

... une t-conorme. Références [Ait04] A. Aït Younes, I. Truck & H. Akdag : Color Image Profil... more

Pattern graph-based image retrieval system combining semantic and visual features

Multimedia Tools and Applications, May 20, 2017

In the literature, several image retrieval approaches that allow mapping between low-level featur... more In the literature, several image retrieval approaches that allow mapping between low-level features and high-level semantics have been proposed. Among these one can cite object recognition, ontologies, and relevance feedback. However, their main limitations concern their high dependence on reliable external resources (existing ontologies, learning sets, etc.) and lack of capacity to combine semantic and visual information and provide relevant results. This paper proposes a system aiming to improve image retrieval results. The proposed system is based on a pattern graph combining semantic and visual features. The idea is (1) to automatically build a modular ontology based on a learning step from textual corpus and terminological resource, (2) to organize visual features in a graph-based model where the combined module and graph represent a unique component called "pattern," and (3) to build a pattern graph. To this end our system has been implemented. The obtained experimental results show that the pattern graph that we propose enables an improvement of retrieval task.

Download

Using genetic algorithms to optimize the number of classification rules in SUCRAGE

Summary form only given. SUCRAGE is a supervised learning method by automatic generation of class... more Summary form only given. SUCRAGE is a supervised learning method by automatic generation of classification rules. The obtained results in generalization using the built rules are satisfactory. However, to be easily interpreted and to allow the explanation of the obtained classification, the rule base size must be reasonable. We propose to optimize the number of rules generated by SUCRAGE using

Trajectory generator design based on the user's intentions for a CMC lower-limbs rehabilitation device

Robotica, Jul 28, 2014

This paper deals with the design of the control structure of a lower-limbs rehabilitation device ... more This paper deals with the design of the control structure of a lower-limbs rehabilitation device in closed muscular chain called Sys-Reeduc. This control structure aims at providing a safe behavior to the user when performing rehabilitation exercises. It is based on two levels. The first level is concerned with the robust trajectory tracking of robotic device and has been the subject of previous studies. Nevertheless, it does not allow, by itself, the user to voluntarily drive the device. Therefore, a trajectory generator constituting the second level is presented in this paper to complete the whole control structure. This high-level control layer is described by a set of dedicated discrete state machines that provide the appropriate sequencing of elementary rehabilitation movements. These elementary movements are dynamically characterized so that clinician may choose the required trajectory parameters to adapt rehabilitation protocols and training to each individual. To realize a complete rehabilitation exercise, the sequence of elementary movements is triggered by thresholds relative to the measurement of the efforts applied by the user on the device. This allows the user to play an active role in its rehabilitation exercises and safely drive the machine at his/her own initiative. The design of the main exercises (isokinetic, isometric, and isotonic) used in the context of lower limbs rehabilitation is described, and simulation results illustrate the effectiveness of the proposed trajectory generator-based control approach.

Download

A linguistic Approach of the Median Aggregator

HAL (Le Centre pour la Communication Scientifique Directe), Sep 1, 2003

International audienc

Towards a Virtual Assistant for Performers and Stage Directors

HAL (Le Centre pour la Communication Scientifique Directe), Jun 1, 2006

In this article, we present the first step of our research work to design a Virtual Assistant for... more In this article, we present the first step of our research work to design a Virtual Assistant for Performers and Stage Directors, able to give a feedback from performances. We use a methodology to automatically construct fuzzy rules in a Fuzzy Rule-Based System that detects contextual emotions from an actor's performance during a show. We collect video data from a lot of performances of the same show from which it should be possible to visualize all the emotions and intents or more precisely "intent graphs". To perform this, the collected data defining low-level descriptors are aggregated and converted into high-level characterizations. Then, depending on the retrieved data and on their distribution on the axis, we partition the universes into classes. The last step is the building of the fuzzy rules that are obtained from the classes and that permit to give conclusions to label the detected emotions.

Download

Modélisation et génération de bases de données géographiques imprécises pour les systèmes relationnels - Extension de F-Perceptory et dérivation automatique de modèles

HAL (Le Centre pour la Communication Scientifique Directe), 2017

Les données géospatiales sont imprécises par nature : leur qualification et leur quantification s... more Les données géospatiales sont imprécises par nature : leur qualification et leur quantification sont soit liées à des approximations, soit à des évaluations subjectives. Il est intéressant de développer des méthodes de conception dédiées. Cela a conduit récemment à la proposition de l'approche F-Perceptory. L'approche n'est cependant appropriée que pour modéliser des espaces homogènes de formes simples ne permettant pas de modéliser des espaces complexes avec des structures composites. Notre proposition étend F-Perceptory et définit un processus de dérivation automatique de modèles permettant de générer la base de données relationnelles. ABSTRACT. Geospatial data are imprecise by nature : their qualification and quantification are either linked to approximations or to subjective assessments. It is important to develop design methods that are dedicated to imprecise spatiotemporal data. This leads to the recent proposal of the F-Perceptory approach. F-Perceptory is appropriate to represent homogeneous spaces

Download

Fuzzy-Possibilistic Classification: Resolution of Initialization Problem

Journal of Advanced Computational Intelligence and Intelligent Informatics, Jan 20, 2009

The methods of automatic classification resulting from the artificial intelligence are generally ... more The methods of automatic classification resulting from the artificial intelligence are generally the consequences of a formalism based on an artificial reasoning quasi similar to that of the human expert. All the approaches of automatic classification developed so far, whether in an exact or approximate context, are dissociated from each other by the membership concept of an object to a class. In this paper, we present a new approach hybrid of unsupervised automatic classification under the C-Means (means of C classes) family. This new approach, based on the fusion of fuzzy and possibility theory and initialized by a membership matrix, allows on the one hand to solve simultaneously the problem of overlapping and coincidence, to reduce the noise effect and on the other hand to accelerate the classification process. The model validation is carried out by the FCM (Fuzzy C-Means), the PCM (Possibilistic C-Means)and the FPCM (Fuzzy-Possibilistic C-Means) for two cases of initialization by using Iris, Textured image and Tight human data basis.

$Research paper thumbnail of A 2D-fractional derivative mask for image feature edge detection$

A 2D-fractional derivative mask for image feature edge detection

Feature extraction is a classic problem of machine vision and image processing. Edges are often d... more Feature extraction is a classic problem of machine vision and image processing. Edges are often detected using integer-order differential operators. In this paper, a one-dimensional digital fractional-order Charef differentiator (1D-FCD) is introduced and extended to 2D by a multi-directional operator. The obtained 2D-fractional differentiation (2D-FCD) is a new edge detection operation. The computed multi-directional mask coefficients are computed in a way that image details are detected and preserved. Experiments on texture images have demonstrated the efficiency of the proposed filter compared to existing techniques.

Actuators and Sensors Faults Detection and Isolation of an Actuated Seat via Nonlinear Multi-observers

HAL (Le Centre pour la Communication Scientifique Directe), 2014

Approximative conjunctions processing by multi-valued logic

We study a knowledge base composed of facts and rules of a linguistic nature. We attribute a ling... more We study a knowledge base composed of facts and rules of a linguistic nature. We attribute a linguistic degree to each proposition which represents its veracity. A problem encountered is the following: “When the premise part of a production rule is a conjunction of several propositions and when we observe each of these propositions with some degree, what global degree should we attribute to the observation, so that we can infer via the rule? We propose a method in order to express the neighbourhood degree between a rule and an observation evaluated on the same scale or on two different scales

Classification of the Car Seats by Detecting the Muscular Fatigue in the EMG Signal

HAL (Le Centre pour la Communication Scientifique Directe), Dec 1, 2005

Abstract The objective of this paper is to evaluate and detect muscular tiredness in natural act... more

A New Approach for the Extraction of Moving Objects

Springer eBooks, 2013

We propose in this paper a background subtraction system for image sequences extracted from fixed... more We propose in this paper a background subtraction system for image sequences extracted from fixed camera using Gaussian Mixture Models (GMM) and the analysis of color histograms. This system can achieve best accuracy than a simple GMM while maintaining the same computational resources. Images extracted from the video will first be divided into several areas of equal size where the behavior of each area is monitored by the analysis of color histograms. For each new frame the color histograms of the zones will be calculated and parts reported to have significant variation in histogram will be updated at the background model. Test carried out show that this approach present best results than a simple GMM. This improvement is important for processing in real time environment.

A Tool for Aggregation with Words

HAL (Le Centre pour la Communication Scientifique Directe), 2009

Download

A Tool for Aggregation with Words

HAL (Le Centre pour la Communication Scientifique Directe), 2009

Download

Image Retrieval using Fuzzy Representation of Colors

HAL (Le Centre pour la Communication Scientifique Directe), 2007

Download

Approximate Reasoning in Supervised Classification Systems

Journal of Advanced Computational Intelligence and Intelligent Informatics, Jul 20, 2006

Apprentissage supervisé et raisonnement approximatif, l'hypothèse des imperfections

HAL (Le Centre pour la Communication Scientifique Directe), 2001

A symbolic Approach for Colorimetric Alterations

HAL (Le Centre pour la Communication Scientifique Directe), Sep 1, 2001

Supervised learning using modifiers: application in colorimetrics

Extraction des couleurs dominantes d’une image

HAL (Le Centre pour la Communication Scientifique Directe), Mar 25, 2007

... une t-conorme. Références [Ait04] A. Aït Younes, I. Truck & H. Akdag : Color Image Profil... more

Pattern graph-based image retrieval system combining semantic and visual features

Multimedia Tools and Applications, May 20, 2017

Download

Using genetic algorithms to optimize the number of classification rules in SUCRAGE

Trajectory generator design based on the user's intentions for a CMC lower-limbs rehabilitation device

Robotica, Jul 28, 2014

Download

A linguistic Approach of the Median Aggregator

HAL (Le Centre pour la Communication Scientifique Directe), Sep 1, 2003

International audienc

Towards a Virtual Assistant for Performers and Stage Directors

HAL (Le Centre pour la Communication Scientifique Directe), Jun 1, 2006

Download

Modélisation et génération de bases de données géographiques imprécises pour les systèmes relationnels - Extension de F-Perceptory et dérivation automatique de modèles

HAL (Le Centre pour la Communication Scientifique Directe), 2017

Download

Fuzzy-Possibilistic Classification: Resolution of Initialization Problem

Journal of Advanced Computational Intelligence and Intelligent Informatics, Jan 20, 2009

$Research paper thumbnail of A 2D-fractional derivative mask for image feature edge detection$

A 2D-fractional derivative mask for image feature edge detection

Actuators and Sensors Faults Detection and Isolation of an Actuated Seat via Nonlinear Multi-observers

HAL (Le Centre pour la Communication Scientifique Directe), 2014

Approximative conjunctions processing by multi-valued logic

Classification of the Car Seats by Detecting the Muscular Fatigue in the EMG Signal

HAL (Le Centre pour la Communication Scientifique Directe), Dec 1, 2005

Abstract The objective of this paper is to evaluate and detect muscular tiredness in natural act... more

A New Approach for the Extraction of Moving Objects

Springer eBooks, 2013

Herman Akdag

Related Authors

Uploads

Papers by Herman Akdag

Log In