Article
Computer Science, Information Systems
Bin Zhang, Bahbibi Rahmatullah, Shir Li Wang, Zhaoyan Liu
Summary: This paper proposes a semi-selective medical image encryption algorithm using enhanced two dimensional Logistic map. The algorithm achieves an ideal compromise between encryption speed and security by fully encrypting the region of interest and semi-encrypting the background region. Simulation results show that the cipher-images have good visual quality, high information entropy, low correlation between adjacent pixels, and uniformly distributed histogram.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Eimad Abusham, Basil Ibrahim, Kashif Zia, Muhammad Rehman
Summary: Biometric authentication systems are more convenient and secure than graphical or textual passwords, but they are susceptible to spoofing attacks. This paper proposes an image encryption scheme integrated into LDA-based face recognition to counter spoofing attacks. The encryption scheme uses XOR pixels substitution and cellular automata for scrambling and a single key for encrypting training and testing datasets. The system achieves high accuracy in classifying encrypted test face images and shows resistance to spoofing attacks.
Article
Automation & Control Systems
Amal H. Alharbi, S. Karthick, K. Venkatachalam, Mohamed Abouhawwash, Doaa Sami Khafaga
Summary: This research focuses on the high-end security applications of face recognition in mobile technologies and computer systems. It proposes an unobtrusive technique to detect face spoofing attacks and proves its effectiveness through experiments.
INTELLIGENT AUTOMATION AND SOFT COMPUTING
(2023)
Article
Computer Science, Information Systems
Dimitrios Kolosov, Vasilios Kelefouras, Pandelis Kourtessis, Iosif Mporas
Summary: Efficient development of on-the-edge Deep Learning applications is complex and challenging, involving exploration of different models, optimization options, and edge devices with trade-offs between accuracy, complexity, and constraints. A case study was conducted using various DL models to detect face masks on five commercial off-the-shelf edge devices, with optimizations made using state-of-the-art frameworks and libraries. Insightful observations were made on the best optimization frameworks, libraries, and devices based on metrics like inference time, efficiency, and value.
Article
Computer Science, Information Systems
Suo Gao, Rui Wu, Xingyuan Wang, Jiafeng Liu, Qi Li, Xianglong Tang
Summary: This paper proposes an encryption method, EFR-CSTP, for face images which efficiently encrypts only the facial information. The method generates secret keys using Hash and recognizes the facial information using HOG. The facial image is then input into EFR-CSTP to generate a ciphertext image using scrambling, STP, and XOR diffusion. The evaluation results show that EFR-CSTP achieves excellent performance and efficiency when compared to other algorithms on the same datasets.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Chengjing Wei, Guodong Li
Summary: In this paper, an improved multi-dimensional hyperchaotic system is proposed and used in the design of selective image encryption. The performance analysis and security analysis demonstrate the reliability and security of the algorithm.
IET IMAGE PROCESSING
(2022)
Article
Physics, Multidisciplinary
Conghuan Ye, Shenglong Tan, Zheng Wang, Binghua Shi, Li Shi
Summary: With the rapid growth of social images being collected on social media platforms, there is an urgent need for secure social image dissemination. The proposed scheme combines the TSH transform with social network analysis to protect the dissemination of social images. The encryption and watermarking techniques used in the scheme provide double protection for the images.
Article
Computer Science, Information Systems
De Rosal Ignatius Moses Setiadi
Summary: This research proposed a dilated hybrid edge detection method for image steganography, which aims to increase the data embedding capacity by expanding the edge area. The results showed that this technique improved the quality and imperceptibility of the stego image.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
H. Ashiba
Summary: This paper proposes a novel cancellable biometric recognition and template protection technique that encrypts data securely based on the properties of graph theory. The technique utilizes the concept of bipartite graphs to generate complex cipher text with a shared key. It applies the Graph First Decomposition Mask (GFH) encoding algorithm on face images, encrypts the resultant map, and uses it to produce the second GFH from the image. The results show that the proposed cancelable biometric technique outperforms other techniques in terms of performance.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Ashish Singh, Kakali Chatterjee, Anish Kumar Singh, Neeraj Kumar
Summary: Mobile-edge computing (MEC) is a new architecture providing services at the network edge, with potential applications in healthcare for remote patient monitoring, diagnosis, and treatment. However, there are security and privacy concerns related to remote data access, including unauthorized access and data leakage, which can make the system inconvenient, untrusted, less suitable, and vulnerable.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Chemistry, Multidisciplinary
Qamar Natsheh, Ana Salagean, Diwei Zhou, Eran Edirisinghe
Summary: This paper introduces a selective encryption approach to protect the security of DICOM images and reduce processing time. It automatically selects regions within medical images using pixel thresholding segmentation technique and applies different encryption algorithms based on their importance. Experimental results show that this approach achieves shorter processing time while maintaining good robustness against various attacks.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Huseyin Bilal Macit, Arif Koyun
Summary: Digital images are widely used, but not all images are significant. Valuable images in fields like banking and finance need to be protected from malicious modifications. This paper proposes an active image forgery detection method that encodes and decodes image edge information. The method not only alerts whether the image is forged or not, but also marks the tampered region. It performs well in detecting geometric attacks, self-copy attacks, and JPEG compression.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Computer Science, Information Systems
S. Kiruthika, V Masilamani
Summary: The tremendous growth of data in social media and other platforms has raised questions about authenticity, particularly in regards to face manipulation and character assassination. Image forgery tools are constantly improving, making it challenging for detection systems. Existing solutions based on deep learning lack reliability and may fail when faced with new forgery tools. This paper proposes a novel method that uses Image Quality Assessment (IQA) based features to detect forged faces. The proposed method achieves a high accuracy of 99% on standard datasets and also discusses the generalization and explainability of the model.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Xun Lin, Shuai Wang, Jiahao Deng, Ying Fu, Xiao Bai, Xinlei Chen, Xiaolei Qu, Wenzhong Tang
Summary: This paper proposes a novel method for image manipulation detection called EMT-Net, which can learn and enhance multiple tampering traces, including noise distribution and visual artifacts. By extracting noise features and visual artifacts, EMT-Net can effectively detect and prevent the risks of tampered images.
PATTERN RECOGNITION
(2023)
Article
Physics, Multidisciplinary
Liming Guo, Jianqing He, Guodong Ye
Summary: This paper proposes an asymmetric image encryption algorithm with plaintext correlation by analyzing various image encryption schemes that use chaotic systems. The algorithm utilizes a new three-dimensional chaotic map and a key acquisition model to enhance the keystream associated with plaintext. Experimental results demonstrate the effectiveness of the proposed algorithm.
Article
Multidisciplinary Sciences
Amin Jula, Elankovan A. Sundararajan, Zalinda Othman, Narjes Khatoon Naseri
Summary: This paper proposes a novel operator for the imperialist competitive algorithm called the color revolution operator (CRO), which significantly improves the algorithm's efficiency. The study applies this improved algorithm to optimize service time-cost in cloud computing service composition, showing that ICACRO is more efficient than traditional algorithms.
Article
Computer Science, Information Systems
Osama A. Khashan, Rami Ahmad, Nour M. Khafajah
Summary: This paper introduces the FlexCrypt scheme, which addresses the challenges of security and efficiency in WSNs by introducing features such as dynamic clustering and automated selection of encryption parameters, as well as providing flexible key management and authentication methods.
Review
Computer Science, Information Systems
Mustafa Tareq, Elankovan A. Sundararajan, Aaron Harwood, Azuraliza Abu Bakar
Summary: Clustering data streams, especially evolving data streams, presents challenges for conventional density grid-based clustering algorithms. In this study, a systematic literature review was conducted to summarize existing grid-based clustering algorithms, their limitations, and the challenges they face in handling evolving data streams. The findings revealed a variety of active research studies on density grid-based clustering.
Article
Chemistry, Analytical
Waleed Alomoush, Osama A. Khashan, Ayat Alrosan, Essam H. Houssein, Hani Attar, Mohammed Alweshah, Fuad Alhosban
Summary: This paper proposes a method to solve FCM problems by improving the balance between exploration and exploitation using an improved global best-guided artificial bee colony algorithm (IABC). The fuzzy clustering algorithm based on PIABC, abbreviated as PIABC-FCM, uses the balancing of PIABC to avoid getting stuck into local optima while searching for the best solution.
Article
Multidisciplinary Sciences
Narjes Khatoon Naseri, Elankovan Sundararajan, Masri Ayob
Summary: In this paper, the performance of the novel heuristic search algorithm called Smart Root Search (SRS) was evaluated for solving service time-cost optimization problems in cloud computing service composition. The SRS algorithm showed significant improvements compared to other algorithms in terms of performance and efficiency. The use of symmetry concepts in dividing the problem search space helped the algorithm to avoid premature convergence and maintain high efficiency in higher-dimensional search spaces. The SRS algorithm has potential applications in various optimization problems.
Article
Telecommunications
Hussein Al-Omaisi, Elankovan A. Sundararajan, Raed Alsaqour, Nor Fadzilah Abdullah, Khairul Azmi Abu Bakar
Summary: This paper proposes a new road-topology-assisted geo-based content discovery scheme called GeoISA to address the issues in vehicle-to-vehicle content discovery in vehicular named data networking. GeoISA utilizes a new mechanism called ISA for Interest forwarding suppression control and a strategy called GIDiS for distance-based TCB Interest forwarding to improve the efficiency of content source and network topology discovery.
VEHICULAR COMMUNICATIONS
(2023)
Article
Computer Science, Information Systems
Osama A. Khashan, Nour M. Khafajah
Summary: With the rapid increase in the number of IoT devices, the need for secure authentication mechanisms that are resource-efficient and scalable has become crucial. Conventional authentication approaches are not suitable for IoT devices due to their expensive cryptographic primitives and lack of scalability. Blockchain technology has been proposed as a solution, but existing blockchain-based authentication methods suffer from high overhead and authentication time delays. This study presents a hybrid centralized and blockchain-based authentication architecture for IoT systems, combining the efficiency of edge servers with the decentralized trust of blockchains. Lightweight cryptographic methods are implemented to achieve efficient authentication while limiting resource consumption. The proposed architecture demonstrates significant improvements in computation cost, execution time, and power consumption compared to traditional authentication schemes.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2023)
Article
Engineering, Chemical
Mohammad Hijjawi, Mohammad Alshinwan, Osama A. Khashan, Marah Alshdaifat, Waref Almanaseer, Waleed Alomoush, Harish Garg, Laith Abualigah
Summary: This paper proposes an improved arithmetic optimization algorithm (AOA) called AOACS based on the cuckoo search algorithm to solve engineering optimization problems. AOACS uses cuckoo search algorithm operators to enhance the exploitation operations of AOA and improve convergence ratio. The performance of AOACS is evaluated using benchmark functions and engineering design problems, and compared to state-of-the-art approaches, demonstrating its superior performance.
Article
Computer Science, Information Systems
Ali Sufyan, Khan Bahadar Khan, Osama A. Khashan, Talha Mir, Usama Mir
Summary: The emergence of 5G network addresses the challenges of rapid growth in mobile subscribers, disruptive IoT-based applications, and immense channel capacity requirements for connecting trillions of devices. It provides significant capacity gain, quality-of-service assurance, higher spectral efficiency, improved device battery life, and massive machine-type communication to meet user demands and tackle the traffic load and resource consumption in 5G applications.
Article
Computer Science, Information Systems
Waleed Alomoush, Osama A. Khashan, Ayat Alrosan, Hani H. Attar, Ammar Almomani, Fuad Alhosban, Sharif Naser Makhadmeh
Summary: The proportion of multimedia traffic in data networks has increased due to advancements in IT. This necessitates addressing challenges in protecting multimedia data, such as preventing unauthorized disclosure, tracking the leak's origin, ensuring no unauthorized alterations, and safeguarding intellectual property. Watermarking is a technique used to transfer secure data over the network and prevent discovery by a third party.
JOURNAL OF CLOUD COMPUTING-ADVANCES SYSTEMS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Osama A. M. Khashan, Nour M. Khafajah, Waleed Alomoush, Mohammad Alshinwan, Sultan Samer Atawneh, Mutasem K. K. Alsmadi
Summary: Securing multimedia data on disk drives is a major concern due to their increasing volumes and security flaws. Existing cryptographic schemes have limitations in terms of computational costs and flexibility. Dynamic encryption file systems can mitigate these limitations by automating encryption operations with higher security. However, most current cryptographic file systems overlook the unique features of multimedia data and vulnerabilities in key management and multi-user sharing.
Article
Computer Science, Information Systems
Osama A. Khashan, Sultan Alamri, Waleed Alomoush, Mutasem K. Alsmadi, Samer Atawneh, Usama Mir
Summary: Recently, technology has made significant advancements and is now being integrated into educational settings to enhance distance learning and promote interaction between learners. The integration of the Internet of Things (IoT) into education can improve the teaching and learning process and broaden the learning context for students. However, it is crucial to protect sensitive learning data during transmission and storage.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Environmental Sciences
Mahmood Safaei, Elankovan A. Sundararajan, Shahla Asadi, Mehrbakhsh Nilashi, Mohd Juzaiddin Ab Aziz, M. S. Saravanan, Maha Abdelhaq, Raed Alsaqour
Summary: Obesity and its complications are a major global issue, and Malaysia ranks sixth among Asian countries in terms of adult obesity. This study aimed to investigate and assess the risk factors associated with obesity and overweight in Malaysia. The findings revealed that lack of physical activity, unhealthy diet, insufficient sleep, genetics, and perceived stress were the most significant risk factors for obesity.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Article
Social Sciences, Interdisciplinary
Mahmoud Alawi, Raed Alsaqour, Maha Abdelhaq, Reem Alkanhel, Baraa Sharef, Elankovan Sundararajan, Mahamod Ismail
Summary: HetVNET is a promising network architecture that combines multiple network technologies for vehicles. We proposed a non-cluster adaptive QoS-aware gateway selection scheme that autonomously selects vehicles as LTE gateways based on network load and QoS requirements.
Proceedings Paper
Computer Science, Theory & Methods
Osama A. Khashan
Summary: Cloud computing allows for convenient storage and sharing of large volumes of data, with on-demand services. Security and protection from attacks are key issues. Proxy re-encryption can enhance data security in the cloud, but may lead to computational burdens and delays. This study proposes a novel parallel workload distribution scheme to improve big data re-encryption processes.
2021 IEEE 11TH ANNUAL COMPUTING AND COMMUNICATION WORKSHOP AND CONFERENCE (CCWC)
(2021)