https://dspace.univ-guelma.dz/jspui/handle/123456789/674 2026-04-07T08:08:25Z https://dspace.univ-guelma.dz/jspui/handle/123456789/18289 Titre: Leveraging AI and IoT for Enhanced Surveillance and Efficient Management of Water Dams Auteur(s): BOUMELIT, YASSAMINE Résumé: Ensuring dam safety requires proactive monitoring of multiple critical risks, including floating debris, potential drowning incidents, and the gradual emergence of structural cracks. In response to these challenges, this work presents an integrated intelligent system based on computer vision and artificial intelligence, designed to detect and classify these hazards in real-time from visual data. The proposed system unifies three complementary detection modules within a single platform: floating object detection, drowning detection, and crack detection. Each module is built upon the YOLOv8 deep learning architecture and trained on dedicated, annotated datasets with tailored preprocessing and optimization techniques. This unified approach enables the system to automatically analyze surveillance imagery, generate accurate alerts, and support decision-making in dam management operations. The experimental results, obtained in a controlled environment, confirm the system’s effectiveness and robustness across all three detection tasks. This work lays the foundation for a future deployable solution to enhance dam safety through intelligent and autonomous monitoring. Keywords: Integrated intelligent system, dam safety, deep learning, YOLOv8, computer vision, floating object detection, drowning detection, crack detection, real-time monitoring. 2025-01-01T00:00:00Z https://dspace.univ-guelma.dz/jspui/handle/123456789/18288 Titre: Toward the creation of a Startup for Electronic Signature and Certificates Services under the Algerian Regulation Auteur(s): BOUSSAID, Alaeddine Résumé: As Algeria undergoes its digital transformation, the establishment of legally recognized and technically secure electronic signature services has become a national imperative. Despite the adoption of Law No. 15-04 and its executive decrees providing a legal framework for electronic signatures, the market still lacks a practical, localized, and standards-compliant solution tailored to the Algerian context. This thesis presents the design, implementation, and evaluation of CertiSafe—a scalable and secure platform for electronic certification and digital signatures aligned with Algerian regulations. CertiSafe serves both as a functional prototype and as a strategic foundation for launching a trusted Certification Authority (CA) capable of serving government, business, and civil society. The platform integrates a React.js frontend, a Node.js backend, and a MySQL database, and relies on OpenSSL and the Node.js crypto module for cryptographic operations. Core functionalities include secure user registration, X.509 certificate issuance, document signing, signature validation, and full certificate lifecycle management. The architecture is modular, allowing for future extensions such as mobile access, multi-signature workflows, secure offline signing, and integration with enterprise systems (ERP). Beyond the technical stack, the thesis addresses regulatory compliance with Algerian laws and compares them to international standards . It also proposes a business-oriented vision, paving the way for a startup capable of bridging legal, technical, and market needs in the realm of digital trust services. This work ultimately contributes to building a sovereign digital trust ecosystem in Algeria, combining legal compliance, robust cryptographic practices, and a user-centered software architecture 2025-01-01T00:00:00Z https://dspace.univ-guelma.dz/jspui/handle/123456789/18287 Titre: Poubelles Intelligentes pour la Reconnaissance des Bouteilles et la Récompense des Utilisateurs Auteur(s): BENABID, Manar Résumé: Given the sharp increase in plastic waste and its negative impact on the environment, it has become essential to develop innovative and effective solutions to address this issue. In this context, this project leverages artificial intelligence technologies to design a smart and portable system capable of automatically detecting and sorting plastic bottles. The system is built around a Raspberry Pi unit integrated with a digital camera and various electronic components such as an electric motor, thermal printer, sensors, and indicator lights. For each bottle inserted, the system captures an image and analyzes it using an AI model trained on the MobileNet neural network — a lightweight and efficient model designed for low-resource embedded devices. A large and balanced dataset of over 70,000 images was collected and used to train and fine-tune the model. After integrating the model into the embedded system, experimental results demonstrated high accuracy and fast performance, making this solution a practical step toward enhancing automated sorting operations in various locations such as public areas or institutions, while also promoting environmental awareness and encouraging users to participate in recycling efforts. 2025-01-01T00:00:00Z https://dspace.univ-guelma.dz/jspui/handle/123456789/18286 Titre: AgriSense: Intelligent Disease Detection in Cereals Using RGB Imaging Auteur(s): HAMOUCHI, HALA Résumé: Modern agriculture faces major challenges, including cereal leaf diseases, reduced crop yields, and the impacts of climate change. To address these issues, we developed AgriSense, an innovative system for the early detection of cereal diseases using deep learning and computer vision. AgriSense is based on a sequential master-slave architecture, where the first model determines whether a cereal plant is healthy or infected, and subsequent models identify the specific disease. The system can detect 11 critical diseases affecting wheat, maize, and rice, including brown rust, yellow rust, septoria, and leaf rust in wheat ; wilting, common rust, and gray leaf spot in maize ; as well as bacterial leaf blight, brown spot, leaf blast, and sheath blight in rice. By enabling fast and accurate diagnosis through a user-friendly interface, AgriSense helps farmers take timely action, reduce crop losses, and promote sustainable agricultural practices. This work highlights the potential of deep learning to revolutionize agri 2025-01-01T00:00:00Z