Article
Computer Science, Information Systems
Kaavya Kanagaraj, G. G. Lakshmi Priya
Summary: This paper proposes feature extraction and selection methods for video event detection, including object structure recognition, motion recognition, and shot boundary detection. By extracting object-based features and performing feature selection using the ranking method, the efficiency of event search is improved. Experimental results demonstrate that this method outperforms other methods in recognizing video events.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Theory & Methods
Zhiyuan He, Jun Zhang, LiaoJun Pang, Eryun Liu
Summary: This paper proposes a novel partial fingerprint verification network based on spatial transformer network (STN) and the local self-attention mechanism. The model can be trained end-to-end and learn multi-level fingerprint features automatically. To alleviate the data annotation work, the model is trained in a self-supervision and domain adaptation manner. The experimental results show that the proposed method achieves state-of-the-art performance and is robust to different types of scanners.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2022)
Article
Computer Science, Information Systems
Mustafa M. Al Rifaee, Mohammad M. Abdallah, Mosa Salah, Ayman M. Abdalla
Summary: Hand veins can be effectively used in biometric recognition due to their robustness and uniqueness. Contact-based hand-based biometric systems have become undesirable during the COVID-19 pandemic, leading to the need for contactless recognition systems and databases. This research contributes by creating a database of hand dorsal vein images obtained contact-free, as well as developing a novel method for extracting rotation invariance features.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Engineering, Electrical & Electronic
Yizhuo Song, Pengyang Zhao, Wenming Yang, Qingmin Liao, Jie Zhou
Summary: Finger vein recognition has been attracting attention for its high security and potential. In this study, we propose EIFNet for finger vein verification, which effectively fuses features extracted from binary vein masks and gray original images. We also develop a finger vein pattern extraction method and a finger vein segmentation dataset THUFVS to improve the accuracy of vein masks.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2023)
Article
Computer Science, Software Engineering
Hongbin Li
Summary: This study introduced the development trend of fingerprint recognition and how to use convolutional neural networks to improve the recognition accuracy of damaged fingerprints; the results showed that the improved CNN method demonstrated higher efficiency and accuracy in the recognition process.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2021)
Article
Computer Science, Artificial Intelligence
Xuefei Yin, Yanming Zhu, Jiankun Hu
Summary: An automated fingerprint recognition system for 3D fingerprints is crucial for biometric security and shows great potential. The proposed method based on RV-guided 3D fingerprint reconstruction and TTP feature extraction effectively addresses the challenges of real-time reconstruction and high-accuracy recognition. Experimental results demonstrate its superiority in both reconstruction and recognition accuracy compared to existing methods, making it suitable for practical applications.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Computer Science, Theory & Methods
Shajunyi Zhao, Dongyuan Ge, Jingfeng Zhao, Wenjiang Xiang
Summary: Fingerprint identification technology is an early, widely applied, and cost-effective biometric technology. It has expanded from traditional criminal investigation to various fields such as e-commerce, attendance, and access control, becoming a popular biometric technology.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Computer Science, Information Systems
Prateek Agrawal, Deepak Chaudhary, Vishu Madaan, Anatoliy Zabrovskiy, Radu Prodan, Dragi Kimovski, Christian Timmerer
Summary: Automated bank cheque verification using image processing aims to provide an alternate methodology for processing bank cheques with minimal human intervention, by acquiring key components through deep learning methods for accurate and precise assessment of handwritten components of bank cheques.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Physics, Multidisciplinary
Wen Liu, Qianqian Cheng, Zhongliang Deng, Mingjie Jia
Summary: This paper proposes a phase feature extraction network C-GCN based on multi-dimensional correlation, aiming to extract new features through convolution and graph convolution layers, and demonstrate superior performance in indoor positioning.
Article
Computer Science, Artificial Intelligence
Jie Gao, Licheng Jiao, Xu Liu, Lingling Li, Puhua Chen, Fang Liu, Shuyuan Yang
Summary: The article proposes a flexible framework called MSDCCN, which is based on multiscale geometric prior knowledge, to improve the feature representation learning process in classification tasks. This framework efficiently aggregates multiresolution scattering and multiscale curvelet features and allows for flexible and dynamic reuse of these features in networks.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Review
Computer Science, Artificial Intelligence
Xiaoting Wu, Xiaoyi Feng, Xiaochun Cao, Xin Xu, Dewen Hu, Miguel Bordallo Lopez, Li Liu
Summary: This paper provides a comprehensive review of the problem of Facial Kinship Verification (FKV), covering various aspects such as problem definition, challenges, applications, benchmark datasets, taxonomy of methods, and state-of-the-art performance. The paper also identifies gaps in current research and suggests potential future research directions.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2022)
Article
Computer Science, Information Systems
Afrooz Arzehgar, Fatemeh Davarinia, Mohammad Mahdi Khalilzadeh
Summary: This study proposed a novel method for the segmentation of brain tissue into CSF, GM, and WM based on anisotropic textural analysis of FLAIR images. The method combines the gray level co-occurrence matrix and curvelet transform, utilizes the Relief method for feature selection, and applies SVM and FCM for pixel label recognition. Compared to other methods, this approach performs better in segmentation results and tracking changes in scan sequences.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Theory & Methods
Shima Jalali, Reza Boostani, Mokhtar Mohammadi
Summary: Accurate gender recognition through fingerprint analysis is crucial for detectives at crime scenes. Preprocessing fingerprint images to extract high quality features aids in gender recognition. Various features were proposed and applied with multiple classifiers to achieve gender recognition.
MULTIDIMENSIONAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Computer Science, Theory & Methods
Zhe Cui, Jianjiang Feng, Jie Zhou
Summary: This study proposes a method for dense registration of fingerprints using an end-to-end network, aiming to improve accuracy and quality. Additionally, a fingerprint mosaicking method based on optimal seam selection is introduced. Experimental results show that the proposed method outperforms previous methods in accuracy and matching accuracy.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2021)
Article
Computer Science, Information Systems
Meennapa Rukhiran, Sethapong Wong-In, Paniti Netinant
Summary: The traditional human proctoring approach to student identity verification can be unreliable and time-consuming. This study proposes using the internet of things to develop flexible biometric recognition systems and evaluates their effectiveness compared to traditional methods. The results show that unimodal facial and fingerprint biometric systems are suitable for student identity verification, while multimodal and semi-multimodal systems offer higher accuracy with shorter processing times and higher costs.
Article
Computer Science, Artificial Intelligence
Rahma Fourati, Boudour Ammar, Javier Sanchez-Medina, Adel M. Alimi
Summary: This article describes an optimized Echo State Network (ESN) with different neural plasticity rules for classifying emotions based on EEG time series. The results show that the ESN with intrinsic plasticity outperforms feature-based methods and has certain advantages compared to other existing methods.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Emna Krichene, Wael Ouarda, Habib Chabchoub, Ajith Abraham, Abdulrahman M. Qahtani, Omar Almutiry, Habib Dhahri, Adel M. Alimi
Summary: The development of a Time series Forecasting System is a major concern for Artificial Intelligence researchers. Existing systems typically evaluate temporal features and analyze data behavior over time, leading to uncertain forecasting accuracy. To address this issue, a novel method called TOREESNN is proposed, which incorporates innovative frameworks, uncertainty prediction, and system integration to achieve higher accuracy compared to state-of-the-art techniques.
APPLIED INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Islem Jarraya, Fatma BenSaid, Wael Ouarda, Umapada Pal, Adel M. Alimi
Summary: This paper presents a new Convolutional Neural Network for Animal Face Detection (CNNAFD) and a new backbone network CNNAFD-MobileNetV2 for animal face detection, along with a new Tunisian Horse Detection Database (THDD). The proposed methods achieve high accuracy on different datasets, demonstrating their effectiveness and competitiveness.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Engineering, Multidisciplinary
Yahia Hamdi, Houcine Boubaker, Besma Rabhi, Abdulrahman M. Qahtani, Fahd S. Alharithi, Omar Almutiry, Habib Dhahri, Adel M. Alimi
Summary: This paper introduces an online handwriting trajectory modeling method using a deep learned recurrent neural network (RNN) to implement the beta-elliptic model. By simulating the beta-elliptical approach, the model is able to limit the calculation time and meet the needs of mobile device users. Experimental results demonstrate the efficiency of the proposed RNN model for online handwriting trajectory modeling.
ENGINEERING SCIENCE AND TECHNOLOGY-AN INTERNATIONAL JOURNAL-JESTECH
(2022)
Article
Computer Science, Artificial Intelligence
Wissem Abbes, Zied Kechaou, Amir Hussain, Abdulrahman M. Qahtani, Omar Almutiry, Habib Dhahri, Adel M. Alimi
Summary: Hybrid cloud platforms offer an attractive solution to organizations interested in implementing integrated private and public cloud applications to meet their profitability requirements. The current work proposes an enhanced binary particle swarm optimization (E-BPSO) algorithm to overcome the shortcomings of the standard BPSO algorithm in dealing with service placement problems. The E-BPSO algorithm outperforms state-of-the-art approaches in terms of both cost and execution time.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Hajer Ghodhbani, Mohamed Neji, Abdulrahman M. Qahtani, Omar Almutiry, Habib Dhahri, Adel M. Alimi
Summary: This paper proposes a flexible person generation system for virtual try-on, which can transfer human appearance across images while preserving texture details and structural coherence. The Dress-up framework is utilized to sequentially interchange garments between images and generate dressing effects of high quality. Extensive evaluations show that the proposed framework outperforms other recent methods and handles a wide range of editing functions with high robustness and effectiveness.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Najwa Kouka, Fatma BenSaid, Raja Fdhila, Rahma Fourati, Amir Hussain, Adel M. Alimi
Summary: This paper proposes a novel algorithm called MaOPSO-CA to address the challenges in many-objective problems. The algorithm utilizes the Inverted Generational Distance (IGD) indicator in innovative ways to enhance selection pressure and improve knowledge sharing. Experimental studies and real-world application demonstrate the effectiveness of the algorithm.
INFORMATION SCIENCES
(2023)
Article
Mathematics, Applied
Adel M. Alimi, Safoua Khelifi, Mohsen Miraoui
Summary: In this paper, we establish the existence, global exponential stability, and uniqueness of measure pseudo almost periodic and automorphic solutions to a class of high-order Hopfield neural networks with delays. The main technique relies on appropriate composition theorems combined with the Banach contraction mapping principle. Three numerical examples are provided to demonstrate the effectiveness of the obtained results.
INTERNATIONAL JOURNAL OF COMPUTER MATHEMATICS
(2023)
Article
Computer Science, Information Systems
Marwa Jabberi, Ali Wali, Bidyut Baran Chaudhuri, Adel M. Alimi
Summary: This paper proposes a method for 3D face alignment of 2D face images in the wild with noisy landmarks. It reconstructs a 3D face model and performs alignment and pose correction using deep learning for face recognition. The proposed method achieves comparable or even better recognition performances compared to the best results reported on popular benchmarks.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Hala Neji, Mohamed Ben Halima, Javier Nogueras-Iso, Tarek M. Hamdani, Abdulrahman M. Qahtani, Omar Almutiry, Habib Dhahri, Adel M. Alimi
Summary: This study proposes a joint technique for super resolution and deblurring based on a Deep Convolutional Neural Network. The proposed model achieves satisfactory results in terms of peak-signal-to-noise ratio, structural similarity index measure, information fidelity criterion, and Visual Information Fidelity metrics, demonstrating its high performance in enhancing low blurry images.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Onsa Lazzez, Abdulrahman M. Qahtani, Abdulmajeed Alsufyani, Omar Almutiry, Habib Dhahri, Vincenzo Piuri, Adel M. Alimi
Summary: This article focuses on analyzing users' visual data to predict their interests and other latent attributes. By analyzing the content of individual images and aggregating image-level information, a pre-trained convolutional neural network is used for feature extraction. Experimental results show that this analysis enhances prediction performance.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Engineering, Biomedical
Najwa Kouka, Rahma Fourati, Raja Fdhila, Patrick Siarry, Adel M. Alimi
Summary: This study presents a channel selection method based on a new Binary Many-Objective Particle Swarm Optimization with Cooperative Agents (BMaOPSO-CA), which performs unsupervised feature learning from clean EEG signals to recognize human emotions. Extensive validation on three different public benchmarks was conducted, highlighting the optimal electrode locations related to emotions and analyzing the relationship between specific brain regions and emotions.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Information Systems
Besma Rabhi, Abdelkarim Elbaati, Houcine Boubaker, Umapada Pal, Adel M. Alimi
Summary: In this paper, a novel multi-lingual word handwriting recovery framework is introduced, based on a convolutional denoising autoencoder with an attention model. The framework achieves high competitive results in offline handwriting recognition.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Multidisciplinary Sciences
Hajer Brahmi, Boudour Ammar, Amel Ksibi, Farouk Cherif, Ghadah Aldehim, Adel M. Alimi
Summary: In this paper, the oscillatory behavior of a new class of memristor-based neural networks with mixed delays is studied, and the existence and uniqueness of the periodic solution of the system are proven using the concept of Filippov solutions. Assumptions are determined to ensure the globally exponentially stability of the solution. Furthermore, the adaptive finite-time complete periodic synchronization problem is investigated, and a new adaptive controller and adaptive update rule are developed using Lyapunov-Krasovskii functional approach. A finite-time complete synchronization condition is established in terms of linear matrix inequalities. Finally, an illustrative simulation is provided to validate the main results.
SCIENTIFIC REPORTS
(2023)
Article
Computer Science, Information Systems
Marwa Jabberi, Ali Wali, Bilel Neji, Taha Beyrouthy, Adel M. Alimi
Summary: In this paper, a deep learning-based method for 3D face recognition is proposed. The method does not rely on using face representation methods as a proxy step for Convolutional Neural Networks (CNNs). Instead, 3D ShapeNets are employed for recognizing faces covering the full 3D shape, along with 3D data augmentation techniques to enlarge datasets. The experimental results demonstrate significant improvement in 3D face recognition performance using deep 3D CNNs like 3D ShapeNets.