Article
Computer Science, Information Systems
N. Sasikaladevi, K. Geetha, A. Revathi
Summary: Digital images used in business, government and medical sectors require security measures to protect against threats. This paper proposes an encryption system called EMOTE, which utilizes binary curves to process sensitive images. The system employs a multilayered structure and binary operations for smaller and faster hardware implementation. Experimental results demonstrate the effectiveness and feasibility of the system in processing sensitive images.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Mousomi Roy, Shouvik Chakraborty, Kalyani Mali
Summary: The rapid evolution in digital communication poses a greater risk of unauthorized access to sensitive data. This study focuses on the security of digital images, proposing a novel image encryption technique that utilizes a binary search tree. The optimization of the binary search tree's structure is achieved using the electromagnetism-like optimization approach, with a focus on optimizing image entropy. The proposed approach also incorporates DNA encoding and bitplane decomposition for added security. Experimental results demonstrate the efficiency of the proposed approach when compared to state-of-the-art methods.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Optics
Alireza A. Arab, Mohammad Javad B. Rostami, Behnam Ghavami
Summary: With the advancement of science and technology, users are increasingly inclined to use visual data in medical and military applications. However, unauthorized access to personal and confidential information has become a serious problem in the digital world. This paper proposes an image encryption algorithm that combines two chaos systems, which has been experimentally proven to be more secure than traditional encryption algorithms.
Article
Optics
Simiao Wang, Qiqi Peng, Baoxiang Du
Summary: This paper presents a color image encryption algorithm that combines a four-dimensional chaotic system with DNA technology. Security analysis shows that the algorithm is able to effectively resist various attacks.
OPTICS AND LASER TECHNOLOGY
(2022)
Article
Engineering, Mechanical
Prabir Kumar Naskar, Surojit Bhattacharyya, Kailash Chandra Mahatab, Krishna Gopal Dhal, Atal Chaudhuri
Summary: This paper proposes an efficient image encryption scheme utilizing permutation and diffusion with chaotic mappings, resulting in an encrypted image with high entropy. The scheme demonstrates sensitivity to both the key and plain-image, and has been rigorously tested for robustness and compared to other schemes, proving its suitability for real-time implementation.
NONLINEAR DYNAMICS
(2021)
Article
Computer Science, Information Systems
Shaista Mansoor, Shabir A. Parah
Summary: This paper proposes a distinctive Hybrid Adaptive Image Encryption (HAIE) scheme that modifies the chaotic system's initial conditions and control parameters using statistical parameters of the plain image. The scheme involves two chaotic maps, logistic map and tent map, to generate pseudo-random sequences. The encrypted image is a hybrid image obtained by permuting half of the plain image using the logistic map and the other half using the tent map. Experimental results show that the scheme has good encryption effect, can resist various attacks, and is highly robust against noise and cropping attacks.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Achraf Daoui, Hicham Karmouni, Omar El Ogri, Mhamed Sayyouri, Hassan Qjidaa
Summary: The study introduces a modified version of the logistic chaotic map to enhance the security of a digital image copyright protection scheme. By combining image zero-watermarking and image encryption, the scheme provides a high level of security for transmitting images over the Internet. An efficient solution using the Sine Cosine Algorithm was introduced to resist geometric attacks on encrypted images, demonstrating superior robustness compared to recent image encryption methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Optics
Yuqiang Dou, Mingzhu Zhang, Haiju Fan, Ming Li
Summary: A hardware-friendly and visually meaningful image encryption strategy is proposed, using a binary measurement matrix and de Bruijn sequence. The plain image is first sparsely represented using discrete wavelet transform (DWT) and its sparse coefficient matrix is obtained. This matrix is then measured using a binary measurement matrix generated by an improved integer logistic chaotic map. The measurement result is scrambled using a de Bruijn state sequence and diffused to obtain a secret image, which is finally hidden in the coefficient matrix of a carrier image using the integer wavelet transform (IWT). Simulation experimental results and comprehensive analyses confirm the effectiveness of this encryption strategy in producing visually meaningful cipher images.
OPTICS AND LASER TECHNOLOGY
(2023)
Article
Computer Science, Information Systems
Manish Kumar, Prateek Gupta
Summary: A new encryption algorithm for medical images based on 1D logistic map and pseudo-random numbers has been proposed, allowing for user control over security level by adjusting encryption rounds. The algorithm inserts pseudo-random rows and columns during encryption to counter attacks, and has been tested for robustness and analyzed for various types of attacks. The algorithm shows efficiency and effectiveness through testing on grayscale and RGB images, with encryption results compared to existing algorithms for practical application.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Omer Kocak, Ugur Erkan, Abdurrahim Toktas, Suo Gao
Summary: Image encryption plays a crucial role in internet-based storing and transferring, and chaotic systems are widely used in encryption schemes. However, existing chaotic maps have limitations in performance and range, and key generation approaches are not optimized. This study proposes an image encryption scheme based on key optimization using the particle swarm optimization algorithm and a novel modular integrated logistic exponential map. The scheme demonstrates excellent encryption results and is faster than existing works.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Mathematics
Shaista Mansoor, Parsa Sarosh, Shabir A. Parah, Habib Ullah, Mohammad Hijji, Khan Muhammad
Summary: In this paper, an adaptive encryption scheme for color images is proposed using Multiple Distinct Chaotic Maps (MDCM) and DNA computing. The scheme employs three distinct chaotic maps to encrypt the individual color channels of the image and adapts the map parameters based on statistical characteristics of the image. The scheme demonstrates high security and resilience against attacks, with low computational costs.
Article
Computer Science, Information Systems
Wenqiang Zhang, Zeyu Zhang, Zhen Liu, Jiao Zhang, Na Ren, Hongjiang Wang, Mingxu Wang, Liming Wang, Yue Zhao
Summary: This paper proposes a color image encryption algorithm based on bit-plane permutation and dynamic DNA technology (CIEA-BPD) using the Logistic map. The algorithm utilizes the Logistic map to generate a permutation sequence for shuffling the color bit planes of three components and combines masking matrices with DNA technology to diffuse the pixel value of the color image. Through classical security analysis, CIEA-BPD demonstrates resistance against various attacks. Experimental results verify that CIEA-BPD can efficiently encrypt a random-like cipher-image with a high degree of security.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Mousomi Roy, Shouvik Chakraborty, Kalyani Mali
Summary: This article proposes a novel evolutionary image encryption approach based on an elitist genetic algorithm, a modified logistic map, and DNA encoding. The approach determines the optimal mask for encryption based on information entropy and significantly improves the encryption performance. Experimental results show the proposed approach produces excellent results and is resilient against different types of attacks. A detailed analysis of the system will help adapt the proposed approach to real-life image communications security.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Jingya Wang, Xianhua Song, Ahmed A. Abd El-Latif
Summary: This paper explores the use of particle swarm optimization (PSO) to select the parameters and initial values of chaotic systems, aiming to increase the entropy of chaotic sequences in image encryption. A new image encryption scheme is proposed, which includes key selection, chaotic sequence preprocessing, block scrambling, expansion, confusion, and diffusion. Experimental tests demonstrate that the proposed scheme has good encryption performance and higher security compared to popular image encryption methods.
Article
Engineering, Electrical & Electronic
Chen Wang, Ye Zhang
Summary: A novel image encryption algorithm based on deep neural network (DNN) is proposed in this paper. The algorithm encrypts the original image directly using the DNN and recovers the original image through the corresponding decryption unit. Experimental results demonstrate the effectiveness of the algorithm for image encryption and its ability to resist various attacks.
Article
Biotechnology & Applied Microbiology
Mehdi Alirezanejad, Rasul Enayatifar, Homayun Motameni, Hossein Nematzadeh
Article
Computer Science, Artificial Intelligence
Mehdi Alirezanejad, Rasul Enayatifar, Homayun Motameni, Hossein Nematzadeh
Summary: The gravitational search algorithm based on learning automata (GSA-LA) is more efficient in finding optimum solutions compared to the original GSA and other evolutionary-based algorithms, by utilizing learning capability to select parameters based on spontaneous reactions.
JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE
(2021)
Article
Optics
Abdorreza Babaei, Homayun Motameni, Rasul Enayatifar
Article
Computer Science, Information Systems
Mehdi Yadollahi, Rasul Enayatifar, Hossein Nematzadeh, Malrey Lee, Jae-Young Choi
JOURNAL OF INFORMATION SECURITY AND APPLICATIONS
(2020)
Article
Computer Science, Information Systems
Sedighe Abasabadi, Hossein Nematzadeh, Homayun Motameni, Ebrahim Akbari
Summary: Feature selection aims to enhance accuracy and evaluation performance by selecting the optimal feature subset in high-dimensional data. Ensemble feature selection method, such as Automatic Thresholding Feature Selection (ATFS), combines various feature selection methods to achieve more robust results. ATFS, along with a new filter method called Sorted Label Interference (SLI), shows superior performance on binary datasets and outperforms individual rankers and existing ensemble methods.
INFORMATION SYSTEMS
(2021)
Review
Automation & Control Systems
Keyvan Golalipour, Ebrahim Akbari, Seyed Saeed Hamidi, Malrey Lee, Rasul Enayatifar
Summary: Clustering aims to discover natural groupings of patterns, points, or objects without a deterministic approach to decide the best method for a given set of input data. Clustering ensemble combines computed solutions of base clustering algorithms to achieve stability and robustness, while clustering ensemble selection chooses a subset of base clustering based on quality and diversity for better performance. This survey covers the historical development of data clustering, basic clustering techniques, clustering ensemble algorithms, and clustering ensemble selection techniques for improved quality and diversity.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Hardware & Architecture
Sedighe Abasabadi, Hossein Nematzadeh, Homayun Motameni, Ebrahim Akbari
Summary: This study presents a new hybrid feature selection method, Ga-rank&rand, which combines a wrapper feature selection algorithm based on genetic algorithm and a proposed filter feature selection method. The method effectively addresses the problem of large number of features in microarray datasets and provides highly accurate solutions at the earlier stages of the genetic algorithm.
JOURNAL OF SUPERCOMPUTING
(2022)
Article
Biology
Hossein Nematzadeh, Jose Garcia-Nieto, Ismael Navas-Delgado, Jose F. Aldana-Montes
Summary: Explainable Artificial Intelligence (CAI) makes complex and opaque models understandable to human users. This paper proposes an Ensemble-based Genetic Algorithm Explainer (EGAE) for melanoma cancer detection, which automatically detects and presents informative sections of the image. Experimental results show that EGAE improves the accuracy of explanation compared to LIME efficiently.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Computer Science, Information Systems
Sadegh Asghari, Hossein Nematzadeh, Ebrahim Akbari, Homayun Motameni
Summary: This paper proposes a new model BC-NMIQ to rank the best features and improves it with the IAMB feature selection method. The experiments on multiple datasets show that BC-NMIQ-IAMB significantly improves the average accuracy of existing binary and multi-class algorithms.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Sandro Hurtado, Hossein Nematzadeh, Jose Garcia-Nieto, Miguel-Angel Berciano-Guerrero, Ismael Navas-Delgado
Summary: In recent years, eXplainable Artificial Intelligence (XAI) has gained attention in data analytics, especially in the field of medical problems. This paper investigates the interpretability of modern XAI methods LIME and SHAP on a Melanoma image classification dataset. The results show that XAI methods have advantages in interpreting the results of Melanoma image classification, with LIME performing better in terms of reproducibility and execution time.
BIOINFORMATICS AND BIOMEDICAL ENGINEERING, PT I
(2022)
Article
Mathematics
Navid Mohseni, Hossein Nematzadeh, Ebrahim Akbari
Summary: This paper presents a method for outlier detection in test samples based on similarity measures, and improves classification accuracy by deleting outliers and using a majority voting function for classifiers enhancement.
INTERNATIONAL JOURNAL OF NONLINEAR ANALYSIS AND APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Hossein Movafegh Ghadirli, Ali Nodehi, Rasul Enayatifar
Summary: This algorithm utilizes mRNA encoding imported into DNA encoding, with permutation of image bits using Arnold's Cat Map and diffusion operations applied to permuted color channels through NCML. The secret key is generated by a simple Logistic map, and parameters and initial values are created using the Double MD5 method. Test results demonstrate the effectiveness of encryption with a sufficiently large key space to withstand common attacks.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)