Article
Computer Science, Artificial Intelligence
Chilukamari Rajesh, Ravichandra Sadam, Sushil Kumar
Summary: This paper proposes an evolutionary approach to the design of 3D CNNs for medical image segmentation. By utilizing the Chameleon Swarm Algorithm for neural network structure and hyperparameter search, the proposed method achieves superior performance and fewer parameters compared to other state-of-the-art models.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Computer Science, Interdisciplinary Applications
Tahereh Hassanzadeh, Daryl Essam, Ruhul Sarker
Summary: This study proposes a new method for developing 3D deep convolutional neural networks by converting 2D images and 2D neuroevolutionary networks into 3D networks. The approach results in high accuracy for 3D medical image segmentation, while saving a significant amount of computational and processing time.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2021)
Article
Multidisciplinary Sciences
Xiaofeng Qu, Jiajun Wang, Xiaoling Wang, Yike Hu, Tianwen Tan, Dong Kang
Summary: Earth-rock dams are vital and costly infrastructure projects. Detecting dam zone boundaries is crucial for safety, but current methods rely heavily on human judgement, which is time-consuming and labor-intensive. To address this issue, a fast boundary detection method is proposed, combining the Otsu algorithm with enhanced Harris hawks optimization. With increased computation speed, the proposed method meets the time requirements of engineering projects. By using particle swarm optimization, the exploration stage of Harris hawks optimization is improved, while a tangent function and chaotic sine map enhance convergence speed and robustness. Application of this method reduces calculation time to 20 s, approximately 18.8% of the original time.
Article
Computer Science, Artificial Intelligence
Enrique J. Carmona, Jose M. Molina-Casado
Summary: This work presents a new methodology for simultaneously segmenting anatomical structures in medical images, specifically targeting the optic disc (OD) and fovea in retinal images. The method utilizes an OD-fovea model and an evolutionary algorithm to achieve competitive segmentation results, with sensitivity and specificity of 0.9072 and 0.9995 for the OD and success rates of 97.3% and 99.0% for the fovea. The average segmentation time per image is 29.35 seconds, making it an efficient approach in image-based computer-aided diagnosis systems.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Engineering, Electrical & Electronic
Cheng Peng, Yikun Liu, Weihua Gui, Zhaohui Tang, Qing Chen
Summary: An improved foam segmentation algorithm based on optimal labeling and edge constraints is proposed in this article, achieving higher accuracy in industrial experiments. The average value and variance of the rand index (RI) demonstrate the effectiveness of the method.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Computer Science, Artificial Intelligence
Zhen Wang, Chunyu Wang, Xianghua Li, Chao Gao, Xuelong Li, Junyou Zhu
Summary: Community structure division is a crucial issue in network data analysis, and algorithms based on Markov chains offer promising solutions for community detection. The MCL algorithm utilizes a dynamic process of updating flow distribution matrix and transition matrix, affecting accuracy and computational cost. A Physarum-inspired relationship among vertices is proposed to enhance transition probability in MCL-based community detection algorithms, showing better computational efficiency and detection performance in experiments.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Tahereh Hassanzadeh, Daryl Essam, Ruhul Sarker
Summary: Developing a Deep Convolutional Neural Network (DCNN) is a challenging research topic that requires extensive efforts and computation. This paper proposes an evolutionary-based framework to find precise and small networks for medical image segmentation and introduces an ensemble model to improve the quality of segmentation. The proposed model outperforms previous methods and achieves better results using fewer trainable parameters.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Wei Wang, Xianpeng Wang, Xiangman Song
Summary: This research proposes a sparse convex surrogate model to guide the evolutionary process of CNN design and a balance strategy between computational resources and accuracy in the selection of network architectures. Experimental results show that the proposed method is competitive in the segmentation of steel microstructure images.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Multidisciplinary Sciences
Muhammad Hanif, Anna Tonazzini, Syed Fawad Hussain, Akhtar Khalil, Usman Habib
Summary: In this paper, a document restoration method is proposed to remove unwanted degradation patterns from color ancient manuscripts. Different color spaces are exploited to highlight spectral differences, and PCA is applied to enhance separation among patterns. Pixel-based segmentation and Gaussian Mixture Model are used to remove interfering classes and inpaint degraded pixels with background texture.
Article
Chemistry, Multidisciplinary
Carlos Capitan-Agudo, Beatriz Pontes, Pedro Gomez-Galvez, Pablo Vicente-Munuera
Summary: Analyzing biological images from the microscope is challenging due to their complexity and three-dimensional shapes. This study introduces an evolutionary algorithm to automatically select the best parameters for 3D cell shape segmentation in curved epithelial tissues. The results confirm the algorithm's proper performance in comparison to manually obtained images.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Chemical
Laith Abualigah, Ali Diabat, Putra Sumari, Amir H. Gandomi
Summary: This paper proposes a new multilevel thresholding approach based on the Evolutionary Arithmetic Optimization Algorithm (AOA) called DAOA, which uses Differential Evolution technique to enhance local research of AOA and achieve better results in multilevel thresholding. The efficiency of the proposed DAOA method is evaluated and compared to other methods, showing superior performance and ranking first in all test cases.
Review
Chemistry, Analytical
Catalina-Lucia Cocianu, Cristian Razvan Uscatu, Alexandru Daniel Stan
Summary: Image registration is a crucial image processing tool for recognition, classification, detection, and analysis tasks. It is widely used in various fields such as remote sensing, computer vision, geophysics, medical image analysis, and surveillance. Nature-inspired algorithms and metaheuristics have emerged as effective alternatives to direct optimization methods in solving the image registration problem. This paper investigates and summarizes state-of-the-art evolutionary-based registration methods, reviewing and comparing algorithms in terms of evolutionary components, fitness function, image similarity measures, and accuracy indexes.
Article
Computer Science, Interdisciplinary Applications
J. Prakash, B. Vinoth Kumar
Summary: Several superpixel segmentation techniques have been developed recently, which can combine perceptually similar pixels to form visually significant entities and reduce the amount of primitives needed for future processing stages. This study thoroughly examines various superpixel algorithms, including watershed-based, graph-based, clustering-based, and energy optimization strategies. It also considers benchmark measures and datasets to create a superpixel benchmark and improve researchers' understanding and application of superpixel segmentation techniques.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Tahereh Hassanzadeh, Daryl Essam, Ruhul Sarker
Summary: Developing a convolutional neural network for medical image segmentation is challenging due to limited labelled data, but using Neuroevolution to create deep attention networks can improve accuracy. This technique, proposed in the paper, automatically establishes networks and achieves superior results in both 2D and 3D image segmentation.
JOURNAL OF DIGITAL IMAGING
(2021)
Article
Computer Science, Information Systems
Laith Abualigah, Nada Khalil Al-Okbi, Mohamed Abd Elaziz, Essam H. Houssein
Summary: This study proposes a method combining the Marine Predators Algorithm and Salp Swarm Algorithm to determine the optimal multilevel threshold image segmentation. The solutions obtained are represented using image histograms, and various standard evaluation measures are employed to assess the effectiveness of the proposed segmentation method. Results indicate that the proposed method outperforms other well-known optimization algorithms in the literature.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Operations Research & Management Science
Angel A. Juan, Peter Keenan, Rafael Marti, Sean McGarraghy, Javier Panadero, Paula Carroll, Diego Oliva
Summary: In the context of simulation-based optimization, this paper reviews recent work related to metaheuristics, matheuristics, simheuristics, biased-randomised heuristics, and learnheuristics for solving complex and large-scale optimization problems in various domains. The paper provides an overview of the main concepts and updated references, and highlights the applications of these hybrid optimization-simulation-learning approaches in solving real-life challenges under dynamic and uncertainty scenarios. A numerical analysis is also included to illustrate the benefits across different application fields. The paper concludes by highlighting open research lines on extending the concept of simulation-based optimization.
ANNALS OF OPERATIONS RESEARCH
(2023)
Article
Computer Science, Artificial Intelligence
J. A. Garcia-Pulido, G. Pajares, S. Dormido
Summary: Unmanned aerial vehicles (UAVs) require additional support during the last phase of landing, for which a cognitive computing-based perception system is proposed in this study. This system utilizes on-board camera and intelligence to recognize the specially designed target, allowing the UAVs to land on the platform. The proposed method outperforms existing strategies, especially in the use of color information, as demonstrated in the test with 800 images captured by a smartphone onboard a quad-rotor UAV.
COGNITIVE COMPUTATION
(2023)
Article
Computer Science, Hardware & Architecture
Angel Casas-Ordaz, Diego Oliva, Mario A. Navarro, Alfonso Ramos-Michel, Marco Perez-Cisneros
Summary: This article introduces a method for determining the threshold values for image segmentation using the Runge Kutta (RUN) optimization algorithm. By combining it with opposition-based learning (OBL), a hybrid algorithm called RUN-OBL is created, which can effectively solve high-dimensional problems. Experimental results demonstrate that the proposed approach performs better in terms of image segmentation and optimization of complex problems.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Computer Science, Information Systems
Simrandeep Singh, Nitin Mittal, Harbinder Singh, Diego Oliva
Summary: Image segmentation is a critical stage in image analysis and pre-processing, where pixels are divided into segments based on threshold values. Multi-level thresholding approaches are more effective than bi-level methods, and a new modified Otsu function is proposed that combines Otsu's between-class variance and Kapur's entropy. Experimental results demonstrate the high efficiency of the modified Otsu method in terms of performance metrics.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Diego Oliva, Noe Ortega-Sanchez, Mario A. Navarro, Alfonso Ramos-Michel, Mohammed El-Abd, Seyed Jalaleddin Mousavirad, Mohammad H. Nadimi-Shahraki
Summary: This paper proposes a combination of the minimum cross-entropy method and the Global-best brain storm optimization algorithm for image segmentation. The method aims to find the best configuration of thresholds by optimizing the minimum cross entropy, and extract regions of interest.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Hardware & Architecture
Ehsan Bojnordi, Seyed Jalaleddin Mousavirad, Mahdi Pedram, Gerald Schaefer, Diego Oliva
Summary: This paper presents a novel MLP training algorithm based on Levy flight distribution, which uses random walks to explore the search space and optimize the performance of neural networks in pattern classification tasks. Experimental results show the superiority of the proposed algorithm compared to other methods.
NEW GENERATION COMPUTING
(2023)
Article
Agricultural Engineering
Parijata Majumdar, Diptendu Bhattacharya, Sanjoy Mitra, Ryan Solgi, Diego Oliva, Bharat Bhusan
Summary: This paper uses XGBoost and BayGA-RF algorithms to predict the irrigation water demand and suitable fertilizer selection for different growth stages of rice. The results show that this method outperforms other methods in terms of prediction accuracy and achieves an accuracy of 98% in predicting suitable fertilizer selection.
PADDY AND WATER ENVIRONMENT
(2023)
Article
Engineering, Biomedical
Santwana S. Gudadhe, Anuradha D. Thakare, Diego Oliva
Summary: Blood vessels in brain tissue can leak or burst, leading to potentially fatal intracranial hemorrhage. This study aims to classify intracranial hemorrhage computed tomography images using texture-based approaches and machine learning classifiers. The results show that the Weber local descriptor performs well for texture code classification.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Review
Engineering, Multidisciplinary
Mohammad Sh. Daoud, Mohammad Shehab, Laith Abualigah, Mohammad Alshinwan, Mohamed Abd Elaziz, Mohd Khaled Yousef Shambour, Diego Oliva, Mohammad A. A. Alia, Raed Abu Zitar
Summary: Chimp Optimization Algorithm (ChOA) is a recent metaheuristic swarm intelligence method that has been widely used for various optimization problems. It has impressive characteristics such as few parameters, no need for derivation information, simplicity, flexibility, scalability, and a capability to balance exploration and exploitation. Due to these advantages, ChOA has gained significant research interest and has been applied in several domains. This review paper provides an overview of research publications using ChOA, including introductory information, discussions on its operations and theoretical foundation, and detailed descriptions of its recent versions and applications. The evaluation of ChOA is also provided. Overall, this review paper is helpful for researchers and practitioners in various fields and provides potential future research directions.
JOURNAL OF BIONIC ENGINEERING
(2023)
Article
Computer Science, Hardware & Architecture
Taymaz Akan, Diego Oliva, Ali-Reza Feizi-Derakhshi, Amir-Reza Feizi-Derakhshi, Marco Perez-Cisneros, Mohammad Alfrad Nobel Bhuiyan
Summary: Image segmentation is a fundamental step in image processing with crucial applications in computer vision, medical imaging, and object recognition. Histogram-based thresholding is a prevalent method for image segmentation. The Battle Royal Optimizer (BRO) is a recent optimization algorithm that shows promise in multilevel image thresholding.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Essam H. Houssein, Mosa E. Hosney, Diego Oliva, Eman M. G. Younis, Abdelmgeid A. Ali, Waleed M. Mohamed
Summary: This paper proposes a wrapper feature selection approach that combines the rat swarm optimization algorithm with genetic operators to improve classification accuracy and reduce the number of features. The approach converts the continuous search space into a discrete space using transfer functions, achieving a balance between local and global search. Experimental results demonstrate the efficiency and effectiveness of the proposed method.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Information Systems
Somnath Chatterjee, Debyarati Saha, Shibaprasad Sen, Diego Oliva, Ram Sarkar
Summary: This paper presents a two-stage facial expression recognition system using thermal images. The first stage utilizes the MobileNet model to extract features from input images. The second stage employs the Moth-flame Optimization algorithm to select the optimal feature subset. The proposed model achieves an accuracy of 97.47% on the thermal image-based facial expressions dataset while using only 29% features generated from the MobileNet model.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Maria Vera, Maria Jose Gomez-Silva, Vicente Vera, Clara I. Lopez-Gonzalez, Ignacio Aliaga, Esther Gasco, Vicente Vera-Gonzalez, Maria Pedrera-Canal, Eva Besada-Portas, Gonzalo Pajares
Summary: An automatic image processing approach based on deep learning and image understanding techniques has been developed to assist dentists in determining bone loss around dental implants. The approach uses a deep learning object detector to roughly identify implants and crowns, and an image understanding process to optimize lines and detect significant points on screw edges. The performance evaluation shows satisfactory results in detecting bone loss.
JOURNAL OF DIGITAL IMAGING
(2023)
Article
Computer Science, Information Systems
Mohamed Abd Elaziz, Abdelghani Dahou, Mohammed Azmi Al-Betar, Shaker El-Sappagh, Diego Oliva, Ahmad O. Aseeri
Summary: Social IoT systems improve the user experience in various applications and the developed QAHA algorithm enhances feature selection using Quantum optimization. Experiments demonstrate the efficiency of QAHA in both UCI and SIoT datasets, showing increased accuracy and decreased feature count.
Article
Computer Science, Information Systems
Laith Abualigah, Diego Oliva, Heming Jia, Faiza Gul, Nima Khodadadi, Abdelazim G. Hussien, Mohammad Al Shinwan, Absalom E. Ezugwu, Belal Abuhaija, Raed Abu Zitar
Summary: A novel hybrid optimization algorithm called IPDOA is proposed in this paper to solve various benchmark functions. By enhancing the search process of PDOA using the primary updating mechanism of DMOA, the proposed method aims to address the main weaknesses of the original methods.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Rui Lv, Dingheng Wang, Jiangbin Zheng, Zhao-Xu Yang
Summary: In this paper, the authors investigate tensor decomposition for neural network compression. They analyze the convergence and precision of tensor mapping theory, validate the rationality of tensor mapping and its superiority over traditional tensor approximation based on the Lottery Ticket Hypothesis. They propose an efficient method called 3D-KCPNet to compress 3D convolutional neural networks using the Kronecker canonical polyadic (KCP) tensor decomposition. Experimental results show that 3D-KCPNet achieves higher accuracy compared to the original baseline model and the corresponding tensor approximation model.
Article
Computer Science, Artificial Intelligence
Xiangkun He, Zhongxu Hu, Haohan Yang, Chen Lv
Summary: In this paper, a novel constrained multi-objective reinforcement learning algorithm is proposed for personalized end-to-end robotic control with continuous actions. The approach trains a single model using constraint design and a comprehensive index to achieve optimal policies based on user-specified preferences.
Article
Computer Science, Artificial Intelligence
Zhijian Zhuo, Bilian Chen, Shenbao Yu, Langcai Cao
Summary: In this paper, a novel method called Expansion with Contraction Method for Overlapping Community Detection (ECOCD) is proposed, which utilizes non-negative matrix factorization to obtain disjoint communities and applies expansion and contraction processes to adjust the degree of overlap. ECOCD is applicable to various networks with different properties and achieves high-quality overlapping community detection.
Article
Computer Science, Artificial Intelligence
Yizhe Zhu, Chunhui Zhang, Jialin Gao, Xin Sun, Zihan Rui, Xi Zhou
Summary: In this work, the authors propose a Contrastive Spatio-Temporal Distilling (CSTD) approach to improve the detection of high-compressed deepfake videos. The approach leverages spatial-frequency cues and temporal-contrastive alignment to fully exploit spatiotemporal inconsistency information.
Review
Computer Science, Artificial Intelligence
Laijin Meng, Xinghao Jiang, Tanfeng Sun
Summary: This paper provides a review of coverless steganographic algorithms, including the development process, known contributions, and general issues in image and video algorithms. It also discusses the security of coverless steganography from theoretical analysis to actual investigation for the first time.
Article
Computer Science, Artificial Intelligence
Yajie Bao, Tianwei Xing, Xun Chen
Summary: Visual question answering requires processing multi-modal information and effective reasoning. Neural-symbolic learning is a promising method, but current approaches lack uncertainty handling and can only provide a single answer. To address this, we propose a confidence based neural-symbolic approach that evaluates NN inferences and conducts reasoning based on confidence.
Article
Computer Science, Artificial Intelligence
Anh H. Vo, Bao T. Nguyen
Summary: Interior style classification is an interesting problem with potential applications in both commercial and academic domains. This project proposes a method named ISC-DeIT, which combines data-efficient image transformer architectures and knowledge distillation, to address the interior style classification problem. Experimental results demonstrate a significant improvement in predictive accuracy compared to other state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Shashank Kotyan, Danilo Vasconcellos Vargas
Summary: This article introduces a novel augmentation technique called Dynamic Scanning Augmentation to improve the accuracy and robustness of Vision Transformer (ViT). The technique leverages dynamic input sequences to adaptively focus on different patches, resulting in significant changes in ViT's attention mechanism. Experimental results demonstrate that Dynamic Scanning Augmentation outperforms ViT in terms of both robustness to adversarial attacks and accuracy against natural images.
Article
Computer Science, Artificial Intelligence
Hiba Alqasir, Damien Muselet, Christophe Ducottet
Summary: The article proposes a solution to improve the learning process of a classification network by providing shape priors, reducing the need for annotated data. The solution is tested on cross-domain digit classification tasks and a video surveillance application.
Article
Computer Science, Artificial Intelligence
Dexiu Ma, Mei Liu, Mingsheng Shang
Summary: This paper proposes a method using neural dynamics solvers to solve infinity-norm optimization problems. Two improved solvers are constructed and their effectiveness and superiority are demonstrated through theoretical analysis and simulation experiments.
Article
Computer Science, Artificial Intelligence
Francesco Gregoretti, Giovanni Pezzulo, Domenico Maisto
Summary: Active Inference is a computational framework that uses probabilistic inference and variational free energy minimization to describe perception, planning, and action. cpp-AIF is a header-only C++ library that provides a powerful tool for implementing Active Inference for Partially Observable Markov Decision Processes through multi-core computing. It is cross-platform and improves performance, memory management, and usability compared to existing software.
Article
Computer Science, Artificial Intelligence
Zelin Ying, Dawei Cheng, Cen Chen, Xiang Li, Peng Zhu, Yifeng Luo, Yuqi Liang
Summary: This paper proposes a novel stock market trends prediction framework called SMART, which includes a self-supervised stock technical data sequence embedding model S3E. By training with multiple self-supervised auxiliary tasks, the model encodes stock technical data sequences into embeddings and uses the learned sequence embeddings for predicting stock market trends. Extensive experiments on China A-Shares market and NASDAQ market prove the high effectiveness of our model in stock market trends prediction, and its effectiveness is further validated in real-world applications in a leading financial service provider in China.
Article
Computer Science, Artificial Intelligence
Hao Li, Hao Jiang, Dongsheng Ye, Qiang Wang, Liang Du, Yuanyuan Zeng, Liu Yuan, Yingxue Wang, C. Chen
Summary: DHGAT1, a dynamic hyperbolic graph attention network, utilizes hyperbolic metric properties to embed dynamic graphs. It employs a spatiotemporal self-attention mechanism and weighted node representations, resulting in excellent performance in link prediction tasks.
Article
Computer Science, Artificial Intelligence
Jiehui Huang, Zhenchao Tang, Xuedong He, Jun Zhou, Defeng Zhou, Calvin Yu-Chian Chen
Summary: This study proposes a progressive learning multi-scale feature blending model for image deraining tasks. The model utilizes detail dilation and texture extraction to improve the restoration of rainy images. Experimental results show that the model achieves near state-of-the-art performance in rain removal tasks and exhibits better rain removal realism.
Article
Computer Science, Artificial Intelligence
Lizhi Liu, Zilin Gao, Yinhe Wang, Yongfu Li
Summary: This paper proposes a novel discrete-time interconnected model for depicting complex dynamical networks. The model consists of nodes and edges subsystems, which consider the dynamic characteristic of both nodes and edges. By designing control strategies and coupling modes, the stabilization and synchronization of the network are achieved. Simulation results demonstrate the effectiveness of the proposed methods.
