Informatique
Permanent URI for this collectionhttps://dspacee.univ-temouchent.edu.dz/handle/123456789/746
Browse
Recent Submissions
Item Plateforme Intelligente de e-Santé pour la Gestion des Rendez-vous Médicaux, du Carnet de Santé Numérique et la Prédiction des Maladies basée sur l’IA – MyVital –(2025) BOUDIEB, Mohamed Anes; RACHEM, Hanane; BOUHALOUAN, DjamilaThis thesis represents the final document of the end-of-study project for obtaining the Master’s degree in Cybersecurity and Artificial Intelligence. It focuses on the study, design, and development of MyVital, an intelligent e-health platform aiming to modernize the Algerian healthcare system. This platform enables users to manage their healthcare journey by simplifying online medical appointment booking, centralizing all their data in a secure digital health record, and assessing disease risks through a predictive module based on artificial intelligence. The ultimate goal of this work is to create a practical and integrated solution that facilitates communication between patients and healthcare professionals, while improving access to care and strengthening prevention. This document provides a solid foundation for future developments of the application, ensuring its continuous improvement and adaptation to user needs.Item Détection et reconnaissance d'objets dans des Images sous marines basée sur le deep learning pour application robotique -Neptunet Net-(2025) ghitri, hamza; Elhadj Mimoune, Mourad; belgrana, Fatima ZohraIn the context of underwater inspection and environmental monitoring, this thesis presents our start-up project “NeptuNet,” a deep learning-based system designed for the automatic identification of underwater gas leaks through marker object detection. The proposed system offers two complementary solutions: a lightweight model, YOLOv5n, suitable for real-time deployment on embedded hardware, and a high-precision model, Faster R-CNN with ResNet-50 backbone, intended for offline analysis. Our approach relies on using computer vision techniques to address the challenges posed by underwater images, such as low contrast and turbidity. The experiments focused on the detection of gas pipes (for navigation assistance), bubble plumes for identifying active leaks, and bubbles considered as a potential leak indicator. Finally, the system was deployed as a web platform, allowing users to test the detection models on their own images or videos.Item Segmentation Sémantique des Images Satellitaires pour la Surveillance Environnementale(2025) LECHLECHE, Maroua; MAKNI, Fatima Zohraa; MESSAOUDI, Mohamed AmineThis project focuses on the semantic segmentation of satellite images using deep lear- ning architectures. The task involves assigning a label to each pixel in an image to automa- tically identify the different elements present in a scene (bare soil, vegetation, urban areas, water bodies, etc.). Following an in-depth study of classical segmentation approaches and deep learning principles, several architectures were implemented and compared. Experiments conducted on an annotated satellite image dataset enabled the evaluation of each model’s performance using metrics such as accuracy, Dice coefficient, and confusion matrix analysis. The results show that models integrating pre-trained backbones provide finer and more generalizable segmentation, especially in visually heterogeneous areas. This work highlights the growing importance of artificial intelligence in the processing of remote sensing data and paves the way for automated decision-support systems in critical fields such as environmental monitoring, security, and risk management.Item Predict Microsatellite Instability (MSI) and Microsatellite Stability (MSS) status in gastrointestinal cancers(2025) Hibi, Lara; Mekadid, Hafsa Amina; Beddad, FatimaIn colorectal cancer (CRC), MSI status is the crucial biomarker that influences treatment approaches, especially immunotherapy. However, the current diagnostic techniques, such as polymerase chain reaction (PCR) and immunohistochemistry (IHC), are time-consuming and variable.This study uses histopathological images to automatically predict the Microsatellite Instability (MSI) and Microsatellite Stability (MSS) status in gastrointestinal cancers using a novel deep learning framework. We suggest an AI-powered approach, Deep learning models to address these issues by combining cutting-edge computer vision methods for nuclear feature segmentation and classification in histopathology slides.The results we achieve shows how deep learning can revolutionise the digital pathology by providing a reliable, accurate, and scalable substitutes for conventional MSI and MSS diagnostics. With implications for enhancing patient outcomes by quicker and more accurate diagnosis, this work is an important breakthrough towards achieving precision oncology.Item Réduction de la consommation d’énergie dans le Cloud Computing.(2025) Benbouha, Djihene; Hassini, Imene; Bouafia, zohirCloud computing is a promising technology that facilitates the execution of applications in various fields. It provides flexible and scalable services on demand with improved quality of service. Ho- wever, this flexibility leads to high energy consumption in data centers, which are the backbone of cloud infrastructure and whose energy needs are growing rapidly. Faced with this significant en- vironmental and economic challenge, reducing the energy impact of this infrastructure is a major priority. In this final year project, we focused on reducing energy consumption in cloud computing, in par- ticular by implementing three main approaches : dynamic voltage and frequency scaling (DVFS), dynamic consolidation of virtual machines, and a combination of these two techniques. We simula- ted these techniques using the SimPy library in Python and analyzed their impact on several key indicators such as energy consumption, execution time, and execution deadline compliance. Our results show that DVFS is suitable for reducing energy consumption (up to 52%), while dyna- mic consolidation optimizes energy consumption by up to 22%, and the combined method achieves a reduction of 49%. This combined approach yielded significant performance gains, particularly for high workloads.Item Segmentation d’images urbaines par Deep Learning(2025) BOUAALAOUI, CHAIMA; SLATNA, AICHA; SAIDISemantic segmentation is a major area of computer vision that aims to assign a precise label to every pixel of an image, thereby enabling a deep and comprehensive understanding of its visual content. Since the rise of deep learning, this field has witnessed significant advances with the introduction of Convolutional Neural Networks (CNNs), which have demonstrated remarkable efficiency in extracting and representing visual features. Among the most popular models is U‐Net, initially designed for precise segmentation in medical contexts, which we have adopted as the foundation of this work. Structural modifications and additional dedicated layers have been introduced to enhance its ability to segment dense visual environments, such as the urban scenes found in the Cityscapes dataset, renowned for its richness and precise annotations. These improvements have significantly increased the model’s precision, while making the learning process more stable, faster, and better suited to the diversity of modern urban environments. This work thus provides a solid methodological basis for designing models adapted to the precise segmentation of complex visual environments.Item Deep Learning et l’Aide au Diagnostic Médical (Application à la Classification d’image médicale)(2025) Bensafi, Basma; Soussi, Amina Nour El Houda; Benomar, Mohamed LamineMedical image classification has seen significant progress in recent years, driven by rapid ad- vancements in medical imaging techniques and the emergence of deep learning methods, which have proven highly effective in the field of computer vision. Deep learning is characterized by its ability to automatically extract features from data without requiring complex preprocessing steps, enabling the construction of rich hierarchical representations of visual information. These developments have enhanced the efficiency of classification systems, making them better suited to handle the visual complexity inherent in many medical conditions, particularly in medical imaging diagnostics, such as diabetic retinopathy classification. This is precisely the focus of our research. Accordingly, we designed an automated system to effectively classify diabetic re- tinopathy using deep learning networks, specifically leveraging transfer learning to ensure high performance and reasonable computational time. We proposed three architectures based on the convolutional models DenseNet121, ViT-B/16, and YOLOv8, which are renowned for their ef- fectiveness on the ImageNet dataset. Our contribution involves classification using two distinct strategies : Direct classification, where the model is applied directly to the images. Cascade classification, where images are classified in stages—first into 2 classes (binary), then 3 classes, and finally 5 classes. The experiments were conducted on two datasets : The first dataset used is APTOS2019, comprising 3,662 images distributed across five (5) distinct classes. The results demonstrated a clear superiority, achieving a maximum accuracy of 96.30%. The second dataset is the DDR (Dataset for Diabetic Retinopathy), consisting of 12,522 images also divided into five (5) distinct classes. The results were satisfactory, with dataset accuracy reaching 93% and 96% respectively.Item Développement d'une application de gestion d'identité décentralisée (DID)(2025) KHETTAB, Oussama; FLIH, Bilel; MEDEDJEL, MansourItem Autonomous AI-Powered App for Blind Users “URE”(2025) DJELIL, Insaf Fatiha; BERROUAYEN, Khedidja; BENDIABDALLAH, Mohammed hakimThe aim of this research was to study the impact of digital technology and artificial intelligence on the social inclusion and autonomy of Algerian visually impaired individuals by developing the URE mobile application. URE offers a novel solution by enabling blind and visually impaired users to interact more independently with their environment through an AI-powered platform incorporating object recognition, text-to-speech reading, and emotion recognition. Running offline and in the local Arabic language, the application is tailored for domestic technology and cultural contexts, enhancing access and online visibility for an underprivileged segment of society. This study was structured into six essential axes. First, we examined the genesis of the idea, objectives, marketing strategy, business strategy, strategic partnerships, and deployment resources. Second, a detailed examination of the project's financial profile was conducted, including income projections and cost estimation. Finally, the development of the application prototype was broken down into primary phases of design, backend and frontend coding, and implementation of AI technologies. In short, URE offers an inclusive and comprehensive solution addressing real issues for blind individuals in Algeria. Through reinforcement of digital visibility and accessibility, the project contributes to building a fairer digital space while encouraging innovation, cooperation, and participation in social and institutional networks. With creative but localized functions, URE demonstrates the potential for assistive technology to shape society toward increased inclusivity.Item Secure e-prescription platform powered by Artificial Intelligence(2025) Kaddache, mohammed el amine; Elmeguenni, nabil; Bendimerad, mohamed; Bendiabdallah, Mohammed HakimThe objective of this study was to design and evaluate the potential of "Rushita," a secure, AI- powered e-prescription platform, to address critical patient safety and efficiency issues within the Algerian healthcare system. Handwritten prescriptions in Algeria frequently lead to errors due to illegibility, posing risks such as incorrect dosages, drug interactions, and allergic reactions, while reliance on paper systems creates inefficiencies and costs. Rushita offers a unique value proposition by providing a fully digital, secure platform that leverages Artificial Intelligence to offer real-time decision support to clinicians, including alerts for interactions and allergies, and access to unified patient histories. The study was conducted across six distinct axes: project presentation, innovative aspects, strategic market analysis, organizational plan, financial viability, and prototype development. Rushita, as Algeria's potential first dedicated e-prescription platform, can significantly reduce medication errors, enhance diagnostic accuracy, improve inter- institutional collaboration, and promote economic and environmental sustainability. In conclusion, Rushita presents a robust and innovative solution aligned with national digital health goals, poised to enhance patient safety, optimize healthcare delivery, and contribute significantly to the modernization of Algeria's healthcare sector.Item Study of Security and Privacy Challenges in Federated Learning Systems(2025) Zingara, mohamed seddik; Younes, ilies; Benaribi, Fethi ImadL'apprentissage fédéré (FL) représente une avancée majeure face à la prolifération des données sur les appareils périphériques, permettant l'entraînement collaboratif de modèles d'apprentissage automatique sans centralisation des données brutes. Cette approche préserve la confidentialité des utilisateurs et réduit les coûts de transfert de données, trouvant des applications cruciales dans des domaines sensibles comme la santé et la finance. Cependant, le partage des mises à jour de modèle expose les systèmes FL à de nouvelles vulnérabilités, notamment les attaques par empoisonnement de modèle et les attaques par inférence, qui compromettent la confidentialité et l'intégrité du système. Cette thèse vise à approfondir la compréhension des principes fondamentaux de l'apprentissage fédéré et à investiguer expérimentalement diverses attaques ciblant tant les aspects de sécurité que de confidentialité des systèmes FL. En matière de sécurité, une attention particulière est portée aux attaques contre les agrégateurs de modèles, tels que FedGA, en analysant leurs vulnérabilités aux manipulations malveillantes qui pourraient nuire à l'intégrité du modèle global. Côté confidentialité, l'étude explore les risques de fuite d'informations sensibles à partir des mises à jour de modèles locaux. Par cette double investigation, la thèse cherche à mettre en lumière les forces et les faiblesses des cadres FL actuels et à contribuer au développement de systèmes d'apprentissage distribués plus robustes et sécurisés.Item Test d’inconsistance du problème de satisfaction de contraintes basé sur la coloration de la microstructure(2020) Mahieddine, Ilham; Bouzid, Radja; Mohamed, Réda Saïdiseveral problems such as coloring of geographic maps, job design... However there are two large families : The first family represents the complete methods which consists in making a route total solution search space, however their complexity increases exponentially with the size of the problem. The second family represents the incomplete methods that make a local exploration in the space of search for solutions. These methods have been used effectively to find solutions to large problems that complete methods cannot solve. These incomplete methods do not prove the inconsistency of a Constraint Satisfaction Problem (CSP) instance. Among the challenges put forward by the CP community of Selman and al in 1997 is to propose efficient incomplete methods to show the inconsistency of Constraint Satisfaction Problem (CSP). Our work consists of taking up and possibly continuing the contribution of Saïdi and Benhamou made in 2008 in which the authors introduce a new incomplete method for the inconsistency test which is based on the notion of dominance between CSPs ( Constraint Satisfaction Problem ) and the staining of the Constraint Satisfaction Problem (CSP) microstructure.Item Logiciel d'Anonymat pour Concours et Examens (Application au Concours de Doctorat de l'Université)(2020) SOUAFI, Khaled; BENOMAR, Mohammed LamineThe main objective of this work is to develop an anonymity application for doctoral competition within a university in an efficient and intelligent way. The application has been designed according to the needs of teachers and managers who are used to manually preparing for this kind of competition. Firstly, we presented the types of codes for anonymity, then we detailed the models and the tools used to carry out this project. Finally we presented our application with some code and screen figures of the application. Our application is simple to use with different parameters, but still needs improvement to support other needs like QR code devices.Item SEGMENTATION ET INTERPRÉTATION D’IMAGES MÉDICALES PAR APPRENTISSAGE AUTOMATIQUE ET SYSTÈME MULTI-AGENTS(2022) Benameur, Habib; Dahmane, Hayat; Belgrana, Fatima ZohraIn an image processing system, segmentation is primordial. Until now, there is no universal image segmentation method. Any technique is effective only for a given type of image, for a given type of application, and in a given context. The main objective of this work is to achieve the semantic segmentation of medical images using machine learning, a category Artificial Intelligence (AI) algorithm. It is precisely about implementing several approaches by taking advantage of the complementarities between different models of Deep Learning (DL). This allows to increase the robustness of the segmentation process. he challenge of this cooperation is based on segmentation by pixel classification in order to extract useful and relevant information. We have proposed four semantic segmentation approaches using different models of the Unet and Resnet50 architectures. An implementation is then done around a multi-agent system (MAS), providing a distribution of expertise and a reduction in execution time on 3D image sequences (slice). Keywords : Medical Imaging,segmentation, Machine Learning (ML), Artificial Intelligence (AI), Deep Learning (DL), Multi-Agent Systems (MAS).Item Modèles de mobilité pour les VANETs dans les villes intelligentes.(2023) ZEROUAL, Abderrahmane; ZOUHIRI, Farouk Salah Eddine; BENZERBADJ, AliThe objective of this study is to understand the functioning of vehicular networks, commonly known as Vehicular Ad-Hoc Networks (VANETs), and evaluate their performance using simulation. We evaluated the performance of VANETs with different sizes using various mobility models. Additionally, we assessed the performance of inter-vehicle communication using three routing protocols : AODV, DSDV, and OLSR. The obtained results are promising. Simulations were conducted using mobility models. VANETs require specific parameters to navigate on the roads, and mobility models are utilized in VANET simulation. These models enable the representation of vehicle movements on the roads and facilitate the evaluation of network performance in different scenarios The objective of this work is to : — Study the parameters involved in mobility models for VANETs. — Study simulation tools for VANET. — Set up NS-2, NS-3, and SUMO simulation programs for example scenarios with mobile devices.Item Implémentation d’un protocole d’authentification et d’identification basée sur une preuve à divulgation nulle de connaissance.(2023) Khaldi, Fayçal Abdelhadi; Zeriguine, Kenza; Bouchakour Errahmani, HichemThis work is part of the graduation project to obtain the Master’s degree in Networks and Data Engineering (RID). We wish through this project to study, model and implement an accreditation and authentication system in a decentralized context based on Zero Knowledge Proofs. In decentralized systems, where user data is not stored on a centralized server, it is crucial to ensure information security and integrity. However, verifying the identity and authenticity of data remains a major challenge. This is why we have chosen to offer an authentication and accreditation method based on Zero Knowledge Proofs. Our approach consists in using advanced cryptographic protocols to allow the verification of information without revealing confidential details, neither on the accreditation, nor on its holder. Thus, users can prove that they have valid information without having to disclose it explicitly, while remaining anonymous, in order to avoid any follow-up or refusal related to their identity. By implementing our verifiable identifiers system, we aim to make credentials more useful on the web.By using Zero Knowledge Proofs, we allow users to have their university degrees verified by potential employers, while ensuring anonymization and secure authentication. To assess the performance of our system, we perform extensive testing and security analysis. The results show that our approach provides relative efficiency in terms of speed and guarantees verifiable credential privacy.Item Les indicateurs locaux de position pour la tolérance aux pannes dans les réseaux de capteurs sans fil(2023) ZAIMI, Rania Dounia Zed; MESSAOUDI, Mohamed AmineIn graph theory, local position indicators are used to measure the degree of centrality or accessibility of the various vertices inside a graph. The relative advantages of the various vertices may vary depending on the criterion used. The main objective of this end-of-studies project is to propose a fault-tolerant solution which consists in exploiting the local position indicators with the aim of guaranteeing the correct operation of the wireless sensor network even after the failure of some of its components.Item Optimization of Constrained Relay Node Deployment using a Metaheuristic(2023) TOUIL, Walid; BENZERBADJ, AliWireless Sensor Networks have emerged as a fascinating and rapidly growing area of research, capturing the attention of numerous researchers across a wide range of scientific disciplines. These networks, which consist of a collection of small, autonomous sensor and relay nodes capable of wirelessly communicating with each other, hold immense value and offer an array of limitless possibilities. In this thesis, our primary focus will revolve around the domain of surveillance. Our objective is to build a two-tierd topology where sensor nodes are deployed on the sensitive site border to cover each point of we aim to minimize the number of relay nodes deployed while ensuring communications between the all sensor nodes and the sink node, in addition to that we strive to minimize the number of hops in the network. The multi-objective combinatorial optimization problem described above has been resolved using Variable Neighborhood Search metaheuristic algorithm and the wighted sum approach. The obtained results are very encouraging since we can resolve a relative big instances of the problem compared to the exact method. on the other hand, the results in terms of optimal number of relay node to deploy and the average hop count are near optimal.Item Utilisation de l’apprentissage profond pour la détection d’objets(2023) MERKACH, Fatima; BENGOUDIFA, Safaa; SAIDI, SamiraItem Etude comparative des algorithmes dédiés à l’ordonnancement dans un environnement de Cloud Computing(2023) MEGROUSSE, Marwa; YEBBAL, Sarra; BOUAFIAThe Cloud Computing presents a promising technology that facilitates the execution of applications from various domains. It provides flexible scalable services based on user demand and offers seamless access to resources- intensive applications. These advantages are only possible through the use of an effective task scheduling technique. In this work, we propose a comparative study between task scheduling algorithms using the CloudSim simulator, considering several performance metrics such as Makespan, cost, energy consumption, and reliability.