Eugène C. Ezin

Training and Research Institute in Computer Science & Institute of Mathematics and Physics

Secure data is a foremost topic in wireless sensor networks since data are transmitted through wireless channels where attackers may get access to critical information. Addressing critical resource-constrained devices of WSN, existing solutions do not consider the specific constraints of WSN in terms of limited resources. We propose a robust encryp...

Wireless sensor networks (WSNs) have several important applications, both in research and domestic use. Generally, their main role is to collect and transmit data from an ROI (region of interest) to a base station for processing and analysis. Therefore, it is vital to ensure maximum coverage of the chosen area and communication between the nodes fo...

The security of data transmitted through wireless sensor networks remains a challenge. These networks are vulnerable to attacks and subject to strong constraints, such as the energy autonomy, the computing capacity and data storage. We study three fearsome attacks easy to execute. These are tampering, replay and sybil attacks which compromise the i...

Maternal mortality is one of the leading causes of death among child-bearing women in Africa. One of the major challenges facing women in rural areas in Africa is delivering their babies safely, thereby curtailing excessive bleeding during birth. This paper focuses on optimizing costs related to delivery of blood using drone technology by determini...

Wireless sensor networks (WSN) are a collection of autonomous nodes with a limited battery life. They are used in various fields such as health, industry, home automation. Due to their limited resources and constraints, WSNs face several problems. One of these problems is the optimal coverage of a observed area. Indeed, whatever the domain, ensurin...

Wireless sensor networks (WSN) are a collection of autonomous nodes with a limited battery life. They are used in various fields such as health, industry, home automation. Due to their limited resources and constraints, WSNs face several problems. One of these problems is the optimal coverage of a observed area. Indeed, whatever the domain, ensurin...

Learning anywhere at any time is the goal of Mobile learning (m-learning). With the proliferation of smartphones, some universities adopt m-learning to facilitate accessibility of documents and online courses for students. Teachers can plan for online assessment and students can join session, answer and see their score directly. During an exam, the...

This paper presents a system, called AmonAI, that predicts the academic performances of students in the LMD system. The approach used allows to establish, for each of the teaching units of a given semester, some estimates of the students results. To achieve this, various machine learning techniques were used. In order to choose the best model for e...

The goal of monitoring patients' vital parameters is to provide better quality health care services, as well as to autonomously and automatically collect and process biological signals and environmental data from patients and unify the actions of medical specialists for these intensive care units, which are very limited in the underdeveloped countr...

Medical surveillance is based on continuous monitoringof the patients vital parameters. The control of vital signs isessential in different cases. Actually, in the case of criticals healthsissues (road or traffic accident; Pregnancy and pregnancy-relatedcomplication like blood pressure, foetal heart rate, etc...) vitalsparameters need to be continu...

Medical surveillance is based on continuous monitoring of the patients vital parameters. The control of vital signs is essential in different cases. Actually, in the case of criticals healths issues (road or traffic accident; Pregnancy and pregnancy-related complication like blood pressure, foetal heart rate, etc...) vitals parameters need to be co...

The explosion of the web has revealed the limitations of traditional information retrieval systems and consequently has favored the emergence of the personalized information retrieval. In this paper, we propose a custom search system based on a dynamic and hybrid modeling of user's profile. We use a hybrid filtering to re-rank the user search resul...

Awale game is one of the famous board games from Africa with many variants and is now played worldwide in various forms. In this paper, we propose an open-source Application Programming Interface (API) for developers to allow an easy implementation of the various variants of Awale as well as artificial intelligence based players. The API is availab...

The optimization of the energy consumed by data centers is a major concern. Several techniques have tried in vain to overcome this issue for many years. In this panoply, predictive approaches start to emerge. They consist in predicting in advance the resource requirement of the Datacenter’s servers in order to reserve their right quantities at the...

Adversarial games are very studied in artificial intelligence. Among these games, there are the mnk-games. An mnk-game is a board game in which two players take turns in placing a piece of their color on an m × n board. The winner is the player who first gets k pieces of his own color in a row; horizontally, vertically, or diagonally. For the resol...

In this paper, we have demonstrated the efficacy of an automatic continuous speech recognition system by mixing fuzzy and neuronal approaches and an acoustic analysis of the sounds of an under-resourced language. The system we propose integrates the modules such as extraction module, segmentation and phoneme recognition modules and whose the core i...

Falls are the major causes of fatal injury for the elderly population. To remedy this, several elderly people monitoring systems with fall detection functionality have been proposed. In this work, we investigate a video-based method of detecting fall incidents from multiple cameras. Our goal is to propose a novel method to detect falls on the floor...

This paper presents an algorithm for syllable segmentation of speech signals based on the calculation of the singularity exponents in each point of the signal combined with Rényi entropy calculation. Rényi entropy, generalization of Shannon entropy, quantifies the degree of signal organization. We then hypothesize that this degree of organization d...

In this article, we propose a mainly-software hardening technique to totally protect unmodified running operating systems on COTS hardware against transient errors in heavily radiation - flooded environment like high altitude space. The technique is currently being implemented in a hypervisor and allows to control the upper layers of the software s...

In this paper, we study the boundary detection in syllable segmentation field. We describe an algorithm proposed for text-independent syllable segmentation. This algorithm provides a performance comparison between the entropies of Shannon, Tsallis and Renyi in an effective detection of beginning-ending points of syllable in a speech signal. The Sha...

Text-independent speech segmentation is a challenging topic in computer-based speech recognition systems. This paper proposes a novel time-domain algorithm based on fuzzy knowledge for continuous speech segmentation task via a nonlinear speech analysis. Short-term energy, zero-crossing rate and the singularity exponents are the time-domain features...

Real-time estimates of a crowd size is a central task in civilian surveillance. In this paper we present a novel system counting people in a crowd scene with overlapping cameras. This system fuses all single view foreground information to localize each person present on the scene. The purpose of our fusion strategy is to use the foreground pixels o...

During this last decade, many works have been done in order to modelize user interests during his interaction with an information retrieval system and particularly on the web. Despite all these efforts, it remains a real challenge to propose a model able not only to learn user's interests implicitly but also by allowing any user to participate at a...

The volume of data on the web grew in recent years. Then it becomes increasingly difficult for a user to access the right information in a short time. However, several works have been carried out with the aim of proposing algorithms to re-rank the user’s search results on the web by taking into account their profile. In this paper, we propose speci...

The detection of moving objects in a video sequence is the first step in an automatic video surveillance system. This work proposes an enhancement of a codebook-based algorithm for moving objects extraction. The proposed algorithm used a perceptual-based approach to optimize foreground information extraction complexity by using a modified codebook...

A great challenge of practical significance in a recent research topic is to develop computer vision system which can automatically recognize a variety of facial expressions. Such an automated system enables to detect faces, analyzes and interprets facial expressions in a scene although the accomplishment of this task is rather strenuous. There are...

In this paper, we propose a fast polygons fusion algorithm to address the problem of moving object detection from overlapping cameras. Once the foreground pixels are detected in each view, we approximate their contours with polygons and project them into the reference plane. After this, we propose an efficient fusion approach to fuse polygons in or...

In automatic videosurveillance system, the first step is the extraction of moving objects. In this work, we investigate a new approach for foreground-background segmentation by using a modified Codebook algorithm. This approach exploit the perceptual information to optimize the detection. It is an adaptive strategy which is proposed to to reduce th...

This paper reports our efforts toward an ASR system for a new under-resourced language (Fongbe). The aim of this work is to build acoustic models and language models for continuous speech decoding in Fongbe. The problem encountered with Fongbe (an African language spoken especially in Benin, Togo, and Nigeria) is that it does not have any language...

In this work, we propose an elderly monitoring system based on a multi-view human posture recognition from overlapping cameras with an application of fall detection. This system is important in order to help elderly people to stay in a secure environment. It classifies person's pose. These postures (standing, crouching, sitting or lying on the floo...

This paper explores the decision fusion for the phoneme recognition problem through intelligent combination of Naive Bayes and Learning Vector Quantization (LVQ) classifiers and feature fusion using Mel-frequency Cepstral Coefficients (MFCC), Relative Spectral Transform—Perceptual Linear Prediction (Rasta-PLP) and Perceptual Linear Prediction (PLP)...

International audience We address in this paper an improved medical monitoring system through an automatic recognition of human activity in Intensive Care Units (ICUs). A multi camera vision system approach is proposed to collect video sequence for automatic analysis and interpretation of the scene. The latter is performed using Hidden Markov Model...

Human activities recognition in videos sequences is a very current research topic being investigated in computer vision. This paper offers an approach for video analysis by exploiting hidden Markov models. We propose an extension of the standard model by integrating three abstraction layers through the management of hierarchical structure and the t...

Les effectifs dans nos universités et de façon spécifique à l'Université d'Abomey-Calavi, deviennent de plus en plus élevés au fil des années académiques. Le besoin de formation est fortement exprimé par les apprenants comme en témoignent les statistiques contraignant les groupes pédagogiques à ne pas respecter le ratio recommandé par l'UNESCO qui...

Mastering the characteristics of the server workload is a key prerequisite in planning and good management of resources within a datacenter. However, for the purpose of research, it is very difficult to find reliable data that can be used to define models adapted to operations in datacenters and tests. Google unwounded a portion of this problem by...

In this paper, we compare three approaches for decision fusion in a phoneme classification problem. We especially deal with decision-level fusion from Naive Bayes and Learning Vector Quantization (LVQ) clas-sifiers that were trained and tested by three speech analysis techniques: Mel-frequency Cepstral Coefficients (MFCC), Relative Spectral Transfo...

In the context of human action recognition from video sequences in the medical environment, a Temporal Belief-based Hidden Markov Model (HMM) is presented. It allows to cope with human action temporality and enables to manage the data uncertainty and the knowledge incompleteness. The system of activity recognition is based on an HMM with explicit s...

In this work, we address the problem of moving object detection from overlapping cameras. We based on homographic transformation of the foreground information from multiple cameras to reference image. We introduce a new algorithm based on Codebook to get each single views foreground information. This method integrates a region based information int...

Background modeling techniques are used for moving object detection in video. Many algorithms exist in the field of object detection with different purposes. In this paper, we propose an improvement of moving object detection based on codebook segmentation. We associate the original codebook algorithm with an edge detection algorithm. Our goal is t...

In speech recognition, phoneme classification has recently gained increased attention. The combination of classifiers has emerged as a reliable method and is used for decision-making by combining individual opinions to produce a final decision. In this study, we propose a novel classifier based on the combination of Naive Bayes and Learning Vector...

Background modeling techniques are used for moving object detection in video. Many algorithms exist in the field of object detection with different purposes. In this paper, we propose an improvement of moving object detection based on codebook segmentation. We associate the original codebook algorithm with an edge detection algorithm. Our goal is t...

We address in this paper an improved medical monitoring system through an automatic recognition of the human activity in Intensive Care Units (ICUs). A multi camera vision system approach is proposed to collect video sequence for automatic analysis and interpreting of the scene. This latter is performed using Hidden Markov Model (HMM) with explicit...

The first texture descriptors are proposed in 1973 by Haralick[1] and in 1975 by Marie Galloway[2] are still used today for image classification or segmentation in various domains. The majority of these features are defined for gray level images. Many papers have proposed different approaches among them, a parallel study of color and gray level tex...

In the context of human action recognition in the video sequences of medical environment, a temporal belief Hidden Markov Model (HMM) is presented. It allows to cope with human action temporality and enables to manage medical data uncertainty and knowledge incompleteness. The whole system of action recognition based on an HMM with explicit state du...

Mapping on the Internet is a growing field with many applications using cartographic data. Available technologies differ marked ly in their co mplexity, cost and openness. This work addresses a three-component Web mapping solution for area management in a town council in Benin. The imp lementation has been done thanks to a geographic database Postg...

and Motivation In recent years, Multi-sensors based-tracking and recognition become a major problems for the computer vision community. This work addresses the problem of recognizing human activities in medical Units Care Intensives, which is an important research issue for building a medical pervasive and smart environment. For this purpose, we pr...

La reconnaissance des situations à risque pour le malade dans les Unités de Soins Intensifs (USIs) médicaux a un intérêt majeur en terme de système d'aide à la décision pour les médecins surveillants. Cet article présente une approche de reconnaissance de scénarii basée sur les Modèles de Markov Cachés temporels (Hidden Markov Model : HMM). Nous ut...

This work addresses the problem of human activities' recognition in Intensives Care, Units(ICUs) which is an important research issue for building a medical pervasive and smart environment. For this purpose, we develop a model for recognizing representative scenarios of risk situations which are archived by selecting video into the Medical Black Bo...

Nowadays, local area networks appear as efficient means to connect servers for xchanging information through a transmission channel between various heterogeneous components. The observed interruption during communication is sometimes caused by connection failures between components. In this paper, we propose an asynchronous message exchange system...

An important step in the analysis and interpretation of video scenes for recognizing scenario is the aberration corrections introduced during the image acquisition in order to provide and correct real image data. This paper presents an approach on distortion correction based on stereoscopic self-calibration from images sequences by using a multi-ca...

Intrusion Detection has become an integral part of the information security process. The cost involved in protecting network resources is often neglected when compared with the actual cost of a successful intrusion, which strengthens the need to develop more powerful intrusion detection systems. Many existing systems for intrusion detection are dev...

Data transmission through a secure channel is important in our daily experience with the objective to ensure that the receiver is the only one authorized and able to discover the message content. In this paper, and based upon the proposed model in [1], we implement a cryptosystem for data transmission. The plaintext message to transmit is first com...

Data transmission through a secure channel requires the attention of many researchers. In this paper, on the basis of an alphabet of ciphers and letters, we propose a model for data transmission through a secure channel. This is achieved at two levels. First we associate each distinct symbol with a probability in the message to transmit. By doing s...

Character classification is known to be one of many basic applications in the field of artificial neural networks (ANN), while data transmission with low size is important in the field of source coding. In this paper, we constructed an alphabet of 36 letters which are encoded with the Huffman algorithm and then classified with a back-propagation Fe...

Feature extraction plays a primary role in pattern recognition classification. Many context-based and problem-based algorithms have been proposed providing good performance in high-quality fingerprint imaging but fail when declining with poor-quality fingerprints. The pyramidal algorithm we present in this paper operates on an image matrix layer fo...