Deep Learning content-based search in media files (images and videos) / Ihssane Khadidja Bendjazia

Public ISBD Titre : Deep Learning content-based search in media files (images and videos) Type de document : document multimédia Auteurs : Ihssane Khadidja Bendjazia, Auteur ; Leila Benarous, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département d'informatique Année de publication : 2024 Importance : 52 p. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Option : Distributed networks, systems, and applications Langues : Anglais Mots-clés : Object Detection  Search  Image  Video  AI  YOLOv8  multimedia Résumé : The rapid growth of digital multimedia content has created a new demand for technologies that not only process but also sort and locate images and videos in the oceans of content. Generally, text-based traditional search engines are sometimes not able to retrieve specific visual content based on visual features, leading to technology development in computer vision for content-based image and video retrieval. This work presents the design, implementation, and evaluation of a user-friendly application, "Search by meaning and visual clues," that utilizes AI and YOLOv8 to offer users content-based search features. Our application allows users to probe multimedia sources using the image or video as input, which removes the laborious text-based queries. By employing modern computer vision, our application aims to produce prompt accurate search results with user friendliness being a top priority. note de thèses : Mémoire de master en informatique Deep Learning content-based search in media files (images and videos) [document multimédia] / Ihssane Khadidja Bendjazia, Auteur ; Leila Benarous, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2024 . - 52 p. + 1 disque optique numérique (CD-ROM). Option : Distributed networks, systems, and applications Langues : Anglais Mots-clés : Object Detection  Search  Image  Video  AI  YOLOv8  multimedia Résumé : The rapid growth of digital multimedia content has created a new demand for technologies that not only process but also sort and locate images and videos in the oceans of content. Generally, text-based traditional search engines are sometimes not able to retrieve specific visual content based on visual features, leading to technology development in computer vision for content-based image and video retrieval. This work presents the design, implementation, and evaluation of a user-friendly application, "Search by meaning and visual clues," that utilizes AI and YOLOv8 to offer users content-based search features. Our application allows users to probe multimedia sources using the image or video as input, which removes the laborious text-based queries. By employing modern computer vision, our application aims to produce prompt accurate search results with user friendliness being a top priority. note de thèses : Mémoire de master en informatique

Réservation

Réserver ce document

Exemplaires

Code-barres	Cote	Support	Localisation	Section	Disponibilité
MF 01-79	MF 01-79	CD	BIBLIOTHEQUE DE FACULTE DES SCIENCES	théses (sci)	Disponible

Deepfakes detection using deep learning / Hadjer Koribaa

Public ISBD Titre : Deepfakes detection using deep learning Type de document : texte manuscrit Auteurs : Hadjer Koribaa, Auteur ; Leila Benarous, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département d'informatique Année de publication : 2022 Importance : 46 p. Format : 30 cm. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Option : Networks, systems and distributed applications (Réseaux,systèmes et applications réparties) Langues : Anglais Mots-clés : Deepfakes  Classification  GANs  ROI  Transfer learning  CNNs  FaceForensics++  ResNet50-V2  Deep learning Résumé : Deepfakes or the hyper-realistic imitation of authentic audio-visual content, are widely spread techniques specially with the use of pre-trained generative adversarial network (GANs) that makes it easier to automatically swap a person's face with another in a video. Due to their devastating impacts on the world, distinguishing between real and deepfake videos has become a fundamental issue. Therefore, automated methods to identify these deepfake videos are required in light of recent public scandals. Much research has been devoted to developing detection methods to reduce the potential negative impact of deepfakes. In this study, we developed an open-source platform called « Deepfake Detection », it presents a new detection technique which consists of two parts: the first part is a binary classification model that can classify videos as fake or real. The second part is another model with different input and output from the first one. It can identify which generation method among these three: FaceSwap, Face2Face and DeepFakes were used to create these deepfakes (categorical classification). We collected our samples from FaceForensics++ dataset. The preprocessing phase was necessary for this study, from frame extraction to face cropping (Region of Interest) and data augmentation. After that we split our data into train and test sets. Next, Convolutional Neural Networks (CNNs) and transfer learning approaches were employed for this work, we implemented Seven CNN-pretrained models in the first part (binary classification), with several trials and different fine-tuning parameters to determine which model is the most suitable for our situation as well as the criteria that influenced our results. The selected CNN-pretrained models are: VGG16, Inception-v3, InceptionResNet-v2, Xception, MobileNet-V2, MobileNet-V3 and ResNet50-V2. Based on the evaluation results, we reached the highest accuracy of 100% with ResNet50-V2 in both parts (the binary and categorical classification). Lastly, we developed our web application, for users to interact with our models. note de thèses : Mémoire de master en informatique Deepfakes detection using deep learning [texte manuscrit] / Hadjer Koribaa, Auteur ; Leila Benarous, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2022 . - 46 p. ; 30 cm. + 1 disque optique numérique (CD-ROM). Option : Networks, systems and distributed applications (Réseaux,systèmes et applications réparties) Langues : Anglais Mots-clés : Deepfakes  Classification  GANs  ROI  Transfer learning  CNNs  FaceForensics++  ResNet50-V2  Deep learning Résumé : Deepfakes or the hyper-realistic imitation of authentic audio-visual content, are widely spread techniques specially with the use of pre-trained generative adversarial network (GANs) that makes it easier to automatically swap a person's face with another in a video. Due to their devastating impacts on the world, distinguishing between real and deepfake videos has become a fundamental issue. Therefore, automated methods to identify these deepfake videos are required in light of recent public scandals. Much research has been devoted to developing detection methods to reduce the potential negative impact of deepfakes. In this study, we developed an open-source platform called « Deepfake Detection », it presents a new detection technique which consists of two parts: the first part is a binary classification model that can classify videos as fake or real. The second part is another model with different input and output from the first one. It can identify which generation method among these three: FaceSwap, Face2Face and DeepFakes were used to create these deepfakes (categorical classification). We collected our samples from FaceForensics++ dataset. The preprocessing phase was necessary for this study, from frame extraction to face cropping (Region of Interest) and data augmentation. After that we split our data into train and test sets. Next, Convolutional Neural Networks (CNNs) and transfer learning approaches were employed for this work, we implemented Seven CNN-pretrained models in the first part (binary classification), with several trials and different fine-tuning parameters to determine which model is the most suitable for our situation as well as the criteria that influenced our results. The selected CNN-pretrained models are: VGG16, Inception-v3, InceptionResNet-v2, Xception, MobileNet-V2, MobileNet-V3 and ResNet50-V2. Based on the evaluation results, we reached the highest accuracy of 100% with ResNet50-V2 in both parts (the binary and categorical classification). Lastly, we developed our web application, for users to interact with our models. note de thèses : Mémoire de master en informatique

Réservation

Réserver ce document

Exemplaires

Code-barres	Cote	Support	Localisation	Section	Disponibilité
MF 01-55	MF 01-55	Thése	BIBLIOTHEQUE DE FACULTE DES SCIENCES	théses (sci)	Disponible

Multimodal-based Pediatric Diseases Diagnosis / Asma Nedjem

Public ISBD Titre : Multimodal-based Pediatric Diseases Diagnosis Type de document : document multimédia Auteurs : Asma Nedjem, Auteur ; Leila Benarous, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département d'informatique Année de publication : 2024 Importance : 52 p. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Specialization : Information System and Decision Langues : Anglais Mots-clés : Pediatric diseases classification  computer vision  ResNet50  LSTM  Multimodal Learning  Models Fusion  android mobile app Résumé : In our project, we aimed to solve the pediatric diseases delayed or mistaken diagnosis cases. For this prototype, we concentrated on only five pediatric diseases classification as a proof of concept. The selected have confusing visual symptoms, they are Chickenpox, Kawasaki, Measles, Roseola and Scarlet fever. For the classification, we use both the images capturing the visual symptoms and textual symptoms representing the internal and external state of the patient. The image dataset was collected from public repositories, while the textual dataset was synthetically constructed. For images disease classification we used Residual Network (ResNet50) and for text disease classification we used Long Short-Term Memory (LSTM). The purpose behind using two models was to create one model by fusing them that can classify diseases based on two types of information. The results for the trained models ResNET50, LSTM and their fusion were good (over 90% of accuracy). Our final application is an android mobile application that uses the fused model to allow doctors to do prompt accurate diagnosis of pediatric diseases. note de thèses : Mémoire de master en informatique Multimodal-based Pediatric Diseases Diagnosis [document multimédia] / Asma Nedjem, Auteur ; Leila Benarous, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2024 . - 52 p. + 1 disque optique numérique (CD-ROM). Specialization : Information System and Decision Langues : Anglais Mots-clés : Pediatric diseases classification  computer vision  ResNet50  LSTM  Multimodal Learning  Models Fusion  android mobile app Résumé : In our project, we aimed to solve the pediatric diseases delayed or mistaken diagnosis cases. For this prototype, we concentrated on only five pediatric diseases classification as a proof of concept. The selected have confusing visual symptoms, they are Chickenpox, Kawasaki, Measles, Roseola and Scarlet fever. For the classification, we use both the images capturing the visual symptoms and textual symptoms representing the internal and external state of the patient. The image dataset was collected from public repositories, while the textual dataset was synthetically constructed. For images disease classification we used Residual Network (ResNet50) and for text disease classification we used Long Short-Term Memory (LSTM). The purpose behind using two models was to create one model by fusing them that can classify diseases based on two types of information. The results for the trained models ResNET50, LSTM and their fusion were good (over 90% of accuracy). Our final application is an android mobile application that uses the fused model to allow doctors to do prompt accurate diagnosis of pediatric diseases. note de thèses : Mémoire de master en informatique

Réservation

Réserver ce document

Exemplaires

Code-barres	Cote	Support	Localisation	Section	Disponibilité
MF 02-71	MF 02-71	CD	BIBLIOTHEQUE DE FACULTE DES SCIENCES	théses (sci)	Disponible

Permission analysis of mobile applications / Imane Hamdi

Public ISBD Titre : Permission analysis of mobile applications Type de document : texte manuscrit Auteurs : Imane Hamdi, Auteur ; Leila Benarous, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département d'informatique Année de publication : 2021 Importance : 53 p. Format : 30 cm. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Option : Networks, systems and distributed applications (Réseaux,systèmes et applications réparties) Langues : Anglais Mots-clés : Mobile applications  Permissions anlysis  Risky permissions  Criminal acts  Privacy violation  Botnet  Data trading Résumé : Mobile phone applications emerged with the introduction of smartphones. Various applications are installed in users’ phones following their interests and satisfying their basic daily needs, app stores are available to allow users to download and install them on their devices. App stores may be a platform for attackers, they can hide malwares and malicious Trojan in applications and upload them, under such assumption mobile phone users can download applications without questioning their level of security .mobile apps require some permission to be granted before being installed. Users have either to grant these accesses to the app and proceed with the installation or reject these permissions and abort the installation of the app. Other permissions can be granted after the installation of the app and users. can disable them. Some of these permissions may be needed for the correct functionality of the application and other times are not even related to the app functioning needs. The risks of the granted permissions may vary from privacy violation and data trading to controlling and using the phone as botnet in criminal acts both cyber and physical crimes. The aim of this thesis is 1) enlighten users about the risks related to permission grants and their misuse cases, 2) investigate the permission accesses of 17 well known phone applications, 3) establish an application that would check the risky permission accesses of installed applications and enables the user to prevent them. note de thèses : Mémoire de master en informatique Permission analysis of mobile applications [texte manuscrit] / Imane Hamdi, Auteur ; Leila Benarous, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2021 . - 53 p. ; 30 cm. + 1 disque optique numérique (CD-ROM). Option : Networks, systems and distributed applications (Réseaux,systèmes et applications réparties) Langues : Anglais Mots-clés : Mobile applications  Permissions anlysis  Risky permissions  Criminal acts  Privacy violation  Botnet  Data trading Résumé : Mobile phone applications emerged with the introduction of smartphones. Various applications are installed in users’ phones following their interests and satisfying their basic daily needs, app stores are available to allow users to download and install them on their devices. App stores may be a platform for attackers, they can hide malwares and malicious Trojan in applications and upload them, under such assumption mobile phone users can download applications without questioning their level of security .mobile apps require some permission to be granted before being installed. Users have either to grant these accesses to the app and proceed with the installation or reject these permissions and abort the installation of the app. Other permissions can be granted after the installation of the app and users. can disable them. Some of these permissions may be needed for the correct functionality of the application and other times are not even related to the app functioning needs. The risks of the granted permissions may vary from privacy violation and data trading to controlling and using the phone as botnet in criminal acts both cyber and physical crimes. The aim of this thesis is 1) enlighten users about the risks related to permission grants and their misuse cases, 2) investigate the permission accesses of 17 well known phone applications, 3) establish an application that would check the risky permission accesses of installed applications and enables the user to prevent them. note de thèses : Mémoire de master en informatique

Réservation

Réserver ce document

Exemplaires

Code-barres	Cote	Support	Localisation	Section	Disponibilité
MF 01-45	MF 01-45	Thése	BIBLIOTHEQUE DE FACULTE DES SCIENCES	théses (sci)	Disponible

Practical guide to networking / Leila Benarous

Public ISBD Titre : Practical guide to networking Type de document : texte imprimé Auteurs : Leila Benarous, Auteur Editeur : Alger : Office des publications universitaires Année de publication : 2023 Importance : 79 p. Format : 30 cm. ISBN/ISSN/EAN : 978-9961-0-2402-7 Langues : Français Mots-clés : Networking  Networks  Réseaux  Réseautage Practical guide to networking [texte imprimé] / Leila Benarous, Auteur . - Alger : Office des publications universitaires, 2023 . - 79 p. ; 30 cm. ISBN : 978-9961-0-2402-7 Langues : Français Mots-clés : Networking  Networks  Réseaux  Réseautage

Réservation

Réserver ce document

Exemplaires

Code-barres	Cote	Support	Localisation	Section	Disponibilité
004.6-106-1	004.6-106-1	Livre externe	BIBLIOTHEQUE DE FACULTE DES SCIENCES	Informatique (SCI)	Disponible
004.6-106-2	004.6-106-2	Livre externe	BIBLIOTHEQUE DE FACULTE DES SCIENCES	Informatique (SCI)	Disponible
004.6-106-3	004.6-106-3	Livre externe	BIBLIOTHEQUE DE FACULTE DES SCIENCES	Informatique (SCI)	Disponible
004.6-106-4	004.6-106-4	Livre externe	BIBLIOTHEQUE DE FACULTE DES SCIENCES	Informatique (SCI)	Disponible
004.6-106-5	004.6-106-5	Livre externe	BIBLIOTHEQUE DE FACULTE DES SCIENCES	Informatique (SCI)	Disponible
004.6-106-6	004.6-106-6	Livre externe	BIBLIOTHEQUE DE FACULTE DES SCIENCES	Informatique (SCI)	Sorti jusqu'au 26/06/2025
004.6-106-7	004.6-106-7	Livre externe	BIBLIOTHEQUE DE FACULTE DES SCIENCES	Informatique (SCI)	Disponible
004.6-106-8	004.6-106-8	Livre externe	BIBLIOTHEQUE DE FACULTE DES SCIENCES	Informatique (SCI)	Disponible
004.6-106-9	004.6-106-9	Livre externe	BIBLIOTHEQUE DE FACULTE DES SCIENCES	Informatique (SCI)	Disponible

Processus concurrents et systèmes parallèles / Leila Benarous

Permalink

Steganographie des images à base DCGAN / Youcef Metabis

Permalink

Steganography using augmented reality / Abderrazak Benarous

Permalink

Système de vote distribué préservant la confidentialité / Katia Inès Cherifi

Permalink

Weapon detection from camera footages using YOLO and SSD models / Meriem Zaoui

Permalink

---

page 1/1