Neurodegenerative disease especially dementia are reported as disease that leads to death, Alzheimer's Disease (AD) is kind dementia that cause progressive and irreversible brain disorder loss which leads to death...
详细信息
Flexible Reconfigurable Manufacturing systems (FRMS) are a radical approach to modern manufacturing, focusing on adaptability and efficiency to respond to dynamic market demands. The fundamental principles and advanta...
详细信息
ISBN:
(数字)9798331540173
ISBN:
(纸本)9798331540180
Flexible Reconfigurable Manufacturing systems (FRMS) are a radical approach to modern manufacturing, focusing on adaptability and efficiency to respond to dynamic market demands. The fundamental principles and advantages of FRMS are discussed in this article. These systems offer a versatile solution to traditional, inflexible manufacturing setups thanks to their modularity, automation, reconfigurability and data-driven decision-making. The key benefits include the ability to adapt to changing production requirements, cost efficiency, improved product quality, and resource optimization. FRMS are found in various industries, such as automotive, electronics, aerospace, and pharmaceuticals, and they meet the needs of customization, precision, and rapid production shifts. FRMS’s future looks promising, as emerging trends like AI integration, sustainability, customization, and interconnectivity are poised to further enhance their capabilities. Manufacturers can excel in an ever-evolving industrial landscape thanks to FRMS’s position in meeting the challenges of an increasingly dynamic global market as they continue to evolve.
作者:
Wiem, GrinaAli, DouikUniversity of Sousse
Networked Objects Control and Communication Systems Laboratory National School of Engineers of Sousse Sousse Technology Center University of Monastir National Engineering School of Monastir Sahloul Belt Road 4054 Rue Ibn Jazzar Monastir5035 Tunisia University of Sousse
Networked Objects Control and Communication Systems Laboratory National School of Engineers of Sousse Sousse Technology Center Sahloul Belt Road 4054 Tunisia
The performance of recognition systems can be significantly affected by various factors, with facial expression poses and lighting changes being the main confounding factors. In order to minimize their impact, we prop...
详细信息
Neurodegenerative disease especially dementia are reported as disease that leads to death, Alzheimer's Disease (AD) is kind dementia that cause progressive and irreversible brain disorder loss which leads to death...
Neurodegenerative disease especially dementia are reported as disease that leads to death, Alzheimer's Disease (AD) is kind dementia that cause progressive and irreversible brain disorder loss which leads to death. AD shows no symptoms in its early stages which makes diagnosing it its beginning a challenge and helpful for doctors as they can slow down its progress in its early stages. Computer-aided approaches such as machine learning which come up with several techniques to detect AD by extracting features from the given image data and use them to build a classifier. Recently, a subcategory of machine learning called deep learning has widely been employed to enhance the medical diagnosis by attempting notable performance. In fact, these approaches avoid the tricky manual feature extraction using Convolutional Neural Network (CNN) considered as a reference in the field of computer vision. This paper proposes a combination of machine-deep learning technics for early diagnosis of AD from positron emission tomography (PET). We first train our CNN on PET images to extract the most relevant features, then we select the most appropriate CNN's level from where the features will be extracted, which will be feed in a second step as input to a Support Vector Machine based classifier (SVM). The proposed approach achieves notable results that exceeded the performance obtained by various existing approaches.
This work proposes a novel generative model, FF-GAN (Frontal Face Generative Adversarial Network), for generating high-quality and diverse frontal faces. FF-GAN utilizes contrastive learning to effectively learn the u...
详细信息
Deep learning has evolved as a discipline that has demonstrated its capacity and usefulness in tackling complicated learning issues as a result of recent improvements in digital technology and the availability of auth...
详细信息
作者:
Wiem, GrinaAli, DouikUniversity of Sousse
Networked Objects Control and Communication Systems Laboratory National School of Engineers of Sousse National Engineering School of Monastir Rue Ibn Jazzar Monastir5035 Tunisia University of Sousse
Networked Objects Control and Communication Systems Laboratory National School of Engineers of Sousse Pôle Technologique de Sousse Route de Ceinture Sahloul Sousse4054 Tunisia
There is a high demand for realistic facial expressions in modern computer graphics and multimedia research. Unfortunately, synthesizing face expressions takes time, effort, and hard labour since high naturalism of fa...
详细信息
Segmentation of stroke lesions holds significant importance in enabling timely diagnosis and subsequent treatment by medical professionals. Automating this process presents technical challenges given the diverse appea...
详细信息
ISBN:
(数字)9798350351484
ISBN:
(纸本)9798350351491
Segmentation of stroke lesions holds significant importance in enabling timely diagnosis and subsequent treatment by medical professionals. Automating this process presents technical challenges given the diverse appearance and evolving nature of lesions, discrepancies among medical experts, limited dataset availability, and the necessity for multiple MRI modalities for comprehensive imaging. This research presents an innovative method called CNL-ResUNet, specifically designed for 2D stroke image segmentation. CNL-ResUNet ingeniously incorporates residual blocks into its architecture, capitalizing on their innate capacity to facilitate seamless information flow across layers, thereby empowering the network to efficiently distill complex patterns from sparse datasets. It integrates customized skip connections to reduce semantic disparities between the encoder and decoder layers. Additionally, the architecture incorporates an innovative Classifier and Localizer (CNL) module, providing supplementary classification and localization insights with enhanced accuracy. By combining these insights with the segmentation results, CNL-ResUNet adeptly reduces the occurrence of false positives and false negatives. To evaluate the model’s performance, experiments were conducted utilizing the SISS dataset from the Ischemic Stroke Lesion Segmentation Challenge 2015. The outcomes from these experiments highlight the superiority of our approach, revealing significant qualitative and quantitative improvements across standard evaluation metrics like the Dice coefficient and Accuracy. Our proposed network demonstrates potential for advancing automated tools in stroke segmentation.
作者:
Grina WiemDouik AliUniversity of Sousse
Networked Objects Control and Communication Systems Laboratory National School of Engineers of Sousse Sousse Technology Center Sahloul Belt Road 4054 University of Monastir National Engineering School of Monastir Rue Ibn Jazzar Monastir 5035 Tunisia University of Sousse
Networked Objects Control and Communication Systems Laboratory National School of Engineers of Sousse Sousse Technology Center Sahloul Belt Road 4054 Tunisia
The performance of recognition systems can be significantly affected by various factors, with facial expression poses and lighting changes being the main confounding factors. In order to minimize their impact, we prop...
详细信息
The performance of recognition systems can be significantly affected by various factors, with facial expression poses and lighting changes being the main confounding factors. In order to minimize their impact, we propose an intriguing method in this study that enables the generation of high-quality images specifically tailored to the target domain. Our objective is to utilize disentangled representation to effectively model the decomposition of data variations and generate neutral facial expression images with frontal posture and adaptive illumination. To achieve this, we incorporate 3D priors in the adversarial learning during the training process, simulating the generation of an analytical 3D face deformation as well as rendering operations. Additionally, we employ contrastive learning to control the disentanglement of the generated faces while preserving the essential properties of facial features. This technique enables us to learn an embedding space in which similar data samples are represented closely, while distinct samples are kept far apart from each other. Furthermore, we conduct an analysis of the learned latent space and introduce several other significant properties that enhance the reinforcement of factor disentanglement. These properties include an imitation learning algorithm, which facilitates the acquisition of meaningful patterns and characteristics.
The recognition of human activities using vision based methods has become a key area of interest in video analytics research. In the past years, several sophisticated deep learning (DL) algorithms have surfaced, desig...
详细信息
ISBN:
(数字)9798350373974
ISBN:
(纸本)9798350373981
The recognition of human activities using vision based methods has become a key area of interest in video analytics research. In the past years, several sophisticated deep learning (DL) algorithms have surfaced, designed to recognize complex human actions depicted in video streams. These approaches have exhibited remarkable competence in tasks related to video analytics. This paper explores the field in greater depth by centering its focus on Sign Language (SL) as a unique and complex form of human activity. SL plays is pivotal in facilitating communication for people who are deaf or hard of hearing, enabling them to interact and engage. The main idea is to recognize Arabic Sign Language especially Tunisian accent, and subsequently translate it into spoken language utilizing artificial intelligence (AI) methods. The goal of this work is to facilitate communication between individuals who are deaf and those who are not. So, this paper presents the comparative results of four deep-learning models: Long Short-Term Memory (LSTM), Bidirectional Long Short-Term Memory (Bi-LSTM), Gated Reccurent Unit (GRU), and Bidirectional Gated Reccurent Unit (Bi-GRU), aiming to improve the recognition accuracy of dynamic signs. The results indicate that the BiLSTM outperforms other variants, showcasing an accuracy of 98%.
暂无评论