School of Computer Science and Informatics

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 20 of 3675
  • ItemOpen Access
    PointCloud-At: Point Cloud Convolutional Neural Networks with Attention for 3D Data Processing
    (MDPI, 2024-10-05) Umar, Saidu; Taherkhani, Aboozar
    The rapid growth in technologies for 3D sensors has made point cloud data increasingly available in different applications such as autonomous driving, robotics, and virtual and augmented reali-ty. This raises a growing need for deep learning methods to process the data. Point clouds are dif-ficult to be used directly as inputs in several deep learning techniques. The difficulty is raised by the unstructured and unordered nature of the point cloud data. So, machine learning models built for images or videos cannot be used directly on point cloud data. Although the research in the field of point clouds has gained high attention and different methods have been developed over the decade, very few research works directly with point cloud data, and most of them con-vert the point cloud data into 2D images or voxels by performing some pre-processing that caus-es information loss. Methods that directly work on point clouds are in the early stage and this af-fects the performance and accuracy of the models. Advanced techniques in classical convolutional neural networks, such as the attention mechanism, need to be transferred to the methods directly working with point clouds. In this research, an attention mechanism is proposed to be added to deep convolutional neural networks that process point clouds directly. The attention module was proposed based on specific pooling operations which are designed to be applied directly to point clouds to extract vital information from the point clouds. Segmentation of the ShapeNet dataset was performed to evaluate the method. The mean intersection over union (mIoU) score of the proposed framework was increased after applying the attention method compared to a base state-of-the-art framework that does not have the attention mechanism.
  • Item
    Assessing web 2D user interface experiences in mixed reality.
    (Elsevier, 2024-05-30) Xing, Yongkang; Fahy, Conor; Shell, Jethro
    Mixed Reality (MR) technologies have the potential to revolutionize how we interact with various fields, such as medicine, education, and communication. However, existing studies have not comprehensively investigated the overall performance of 2D user interfaces (UIs) in 3D spaces. There are gaps and questions that have not been properly addressed in the transition from 2D to 3D UIs. To investigate this, we design an experiment with 80 participants to evaluate the 2D UI user experience on MR platforms. Our study reveals that compared with desktop devices, the website user experience on MR platforms leads to poorer learning performance. One-to-one interviews with participants reveal issues with both the hardware field of view and color definition, as well as the UI. Based on these findings, we propose that a generalized and optimized 3D UI would reduce control difficulties and improve the learning experience provided by MR platforms. Our study provides critical data that can be used to enhance 3D UIs on MR platforms.
  • Item
    Towards Refined Autism Screening: A Fuzzy Logic Approach with a Focus on Subtle Diagnostic Challenges
    (MDPI, 2024-06-28) Smith, Philip; Greenfield, Sarah
    This study explores the creation and testing of a Fuzzy Inferencing System for automating preliminary referrals for autism diagnosis, utilizing membership functions aligned with the Autism Quotient 10-item questionnaire. Validated across three distinct datasets, the system demonstrated perfect accuracy in deterministic settings and an overall accuracy of 92.91% in a broad fuzzy dataset. The use of Fuzzy Logic reflects the complex and variable nature of autism diagnosis, suggesting its potential applicability in this field. While the system effectively categorized clear referral and non-referral scenarios, it faced challenges in accurately identifying cases requiring a second opinion. These results indicate the need for further refinement to enhance the efficiency and accuracy of preliminary autism screenings, pointing to future avenues for improving the system’s performance. The motivation behind this study is to address the diagnostic gap for high-functioning adults whose symptoms present in a more neurotypical manner. Many current deep learning approaches for diagnosing autism focus on quantitative datasets like fMRI and facial expressions, often overlooking behavioral traits. However, autism diagnosis still heavily relies on long histories and multi-stakeholder information from parents, teachers, doctors and behavioral experts. This research addresses the challenge of creating an automated system that can handle the nuances and variability inherent in ASD symptoms. The theoretical innovation lies in the novel application of Fuzzy Logic to interpret these subtle diagnostic indicators, providing a more systematic approach compared to traditional methods. By bridging the gap between subjective clinical evaluations and objective computational techniques, this study aims to enhance the preliminary screening process for ASD.
  • Item
    eXtended Reality of socio-motor interactions: Current Trends and Ethical Considerations for Mixed Reality Environments Design
    (ACM, 2023-10-09) Ayache, Julia; Bieńkiewicz, Marta; Richardson, Kathleen; Bardy, Benoit
    Social interactions are multi-modal, and their translation into virtuous and smooth interactions in digital, hybrid reality spaces constitute a technological challenge with profound but often dismissed ethical considerations (such as harmful consequences). In particular, the increasing reliance on algorithms (i.e., artificial intelligence) to emulate “seamless” communication patterns between human and artificial agents calls for caution in designing such systems. This short paper reviews current trends in rendering hyper-realistic human bodies and motor behaviors, focusing on the ethical issues of manipulating kinematics in human-to-human and human-to-artificial agent interactions.
  • ItemOpen Access
    Adapting genetic algorithms for multifunctional landscape decisions: a theoretical case study on wild bees and farmers in the UK
    (Wiley, 2024-09-19) Knight, Ellen; Balzter, Heiko; Breeze, Tom; Brettschneider, Julia; Girling, Robbie; Hagen-Zanker, Alex; Image, Mike; Johnson, Colin; Lee, Christopher; Lovett, Andrew; Petrovskii, Sergei; Varah, Alexa; Whelan, Mick; Yang, Shengxiang; Gardner, Emma
    1. Spatial modelling approaches to aid land-use decisions which benefit both wildlife and humans are often limited to the comparison of pre-determined landscape scenarios, which may not reflect the true optimum landscape for any end-user. Furthermore, the needs of wildlife are often under-represented when considered alongside human financial interests in these approaches. 2. We develop a method of addressing these gaps using a case-study of wild bees in the UK, an important group whose declines may adversely affect both human economies and surrounding ecosystems. By combining the genetic algorithm NSGA-II with a process-based pollinator model which simulates bee foraging and population dynamics, Poll4pop, we ‘evolve’ a typical UK agricultural landscape to identify optimum land cover configurations for three different guilds of wild bee. These configurations are compared to those resulting from optimisations for farm income alone, as well as optimisations that seek a compromise between bee populations and farm income objectives. 3. We find that the land cover proportions in landscapes optimised for each bee guild reflect their nesting habitat preferences rather than foraging preferences, highlighting a limiting resource within the study landscape. The spatially explicit nature of these optimised landscapes illustrates how improvement for a given target species may be limited by differences between their movement range and the scale of the units being improved. Land cover composition and configuration differ significantly in landscapes optimised for farm income and bee population growth simultaneously and illustrate how human agents are required to compromise much more when the multifaceted nature of biodiversity is recognised and represented by multiple objectives within an optimisation framework. Our methods provide a way to quantify the extent to which real-life landscapes promote or compromise objectives for different landscape end-users. 4. Our investigation suggests that optimisation set-up (decision-unit scales, traditional choice of a single biodiversity metric) can bias outcomes towards human-centric solutions. It also demonstrates the importance of representing the individual requirements of different actors with different landscape-level needs when using genetic algorithms to support biodiversity-inclusive decision-making in multi-functional landscapes.
  • ItemEmbargo
    Learning to search promising regions by space partitioning for evolutionary methods
    (Elsevier, 2024-09-11) Xia, Hai; Li, Changhe; Tan, Qingshan; Zeng, Sanyou; Yang, Shengxiang
    To alleviate the premature, many evolutionary computation algorithms try to balance the exploitation and exploration by controlling the population diversity. However, randomly diversifying a population cannot always guarantee that an algorithm exploits or explores promising regions. To address this issue, a general framework is proposed in this paper for learning promising regions that are made up of subspaces to guide where to exploit and explore by two reinforcement learning systems. The learning mechanism is as follows: (1) To enhance the efficiency of exploitation, an exploitative reinforcement learning system is constructed to estimate the exploitative potential values of subspaces. Accordingly, basins of attraction are approximated by clustering subspaces and historical solutions are selected within the same basin of attraction to generate new solutions. (2) To efficiently explore the solution space, an explorative reinforcement learning system is established to estimate the explorative potential values of subspaces. Accordingly, algorithms are guided to explore subspaces with higher explorative potential values, promoting the discovery of unexploited promising basins of attraction. The framework is implemented into three conventional evolutionary algorithms, and the mechanism and effectiveness of the implemented algorithms are investigated by comprehensive experimental studies. The experimental results show that the proposed algorithms have competitive performances over the other twelve popular evolutionary algorithms.
  • Item
    Assessment of Dataset Scalability for Classification of Black Sigatoka in Banana Crops Using UAV-Based Multispectral Images and Deep Learning Techniques
    (MDPI, 2024-09-19) Gongora, Mario Augusto; Linero-Ramos, R.; Parra-Rodríguez, C.; Espinosa-Valdez, A.; Gómez-Rojas, J.
    This paper presents an evaluation of different convolutional neural network (CNN) architectures using false-colour images obtained by multispectral sensors on drones for the detection of Black Sigatoka in banana crops. The objective is to use drones to improve the accuracy and efficiency of Black Sigatoka detection to reduce its impact on banana production and improve the sustainable management of banana crops, one of the most produced, traded, and important fruits for food security consumed worldwide. This study aims to improve the precision and accuracy in analysing the images and detecting the presence of the disease using deep learning algorithms. Moreover, we are using drones, multispectral images, and different CNNs, supported by transfer learning, to enhance and scale up the current approach using RGB images obtained by conventional cameras and even smartphone cameras, available in open datasets. The innovation of this study, compared to existing technologies for disease detection in crops, lies in the advantages offered by using drones for image acquisition of crops, in this case, constructing and testing our own datasets, which allows us to save time and resources in the identification of crop diseases in a highly scalable manner. The CNNs used are a type of artificial neural network widely utilised for machine training; they contain several specialised layers interconnected with each other in which the initial layers can detect lines and curves, and gradually become specialised until reaching deeper layers that recognise complex shapes. We use multispectral sensors to create false-colour images around the red colour spectra to distinguish infected leaves. Relevant results of this study include the construction of a dataset with 505 original drone images. By subdividing and converting them into false-colour images using the UAV’s multispectral sensors, we obtained 2706 objects of diseased leaves, 3102 objects of healthy leaves, and an additional 1192 objects of non-leaves to train classification algorithms. Additionally, 3640 labels of Black Sigatoka were generated by phytopathology experts, ideal for training algorithms to detect this disease in banana crops. In classification, we achieved a performance of 86.5% using false-colour images with red, red edge, and near-infrared composition through MobileNetV2 for three classes (healthy leaves, diseased leaves, and non-leaf extras). We obtained better results in identifying Black Sigatoka disease in banana crops using the classification approach with MobileNetV2 as well as our own datasets.
  • ItemEmbargo
    Improved Attentive Pairwise Interaction (API-Net) for Fine-Grained Image Classification
    (IEEE, 2022-01-26) Yet, Ong Zu; Rassem, Taha H.; Rahman, Md Arafatur; Rahman, M. M.
    Fine-grained classification is a challenging problem as one has to deal with a similar class of objects but with various types of variations. For more elaboration, they are almost similar and have subtle differences, and are confusing. In this study, aircraft will be the fine-grained object to be focused on. Aircraft which has almost similar shapes and patterns can be hardly recognized even for humans, especially those who haven not gone through any training. In recent years, a lot of proposed methods addressed to solve the difficulties in fine-grained problems by learning contrastive clues from an image. This study aims to increase the accuracy of the Attentive Pairwise Interaction Network (API-Net) by introducing data augmentation into the network structure. Some of the previous studies proved that data augmentation does help improve a network. So, this study is going to modify the API-Net with different data augmentation settings. In this study, various settings have been introduced to the API-Net. Several experiments had been done with a simple modification where a portion of the train dataset’s images will randomly convert into greyscale images. These settings are, only brightness & contrast 0.2, only grayscale 0.3, only grayscale 0.5, brightness & contrast 0.2 with grayscale 0.3, and brightness & contrast 0.2 with grayscale 0.5. As a result, the proposed modification achieved with 92.74% with brightness & contrast 0.2, 92.80% on brightness & contrast 0.2 with grayscale 0.5, and 92.86% on brightness & contrast 0.2 with grayscale 0.3. While grayscale 0.3 alone achieve 93.25% and grayscale 0.5 alone achieve 93.46% compared with the original results which reached 92.77%.
  • ItemOpen Access
    Efficient Authentication Scheme for 5G-Enabled Vehicular Networks Using Fog Computing
    (MDPI, 2023-03-28) Al-Mekhlafi, Zeyad Ghaleb; Al-Shareeda, Mahmood A.; Manickam, Selvakumar; Mohammed, Badiea Abdulkarem; Alreshidi, Abdulrahman; Alazmi, Meshari; Alshudukhi, Jalawi Sulaiman; Alsaffar, Mohammad; Rassem, Taha H.
    Several researchers have proposed secure authentication techniques for addressing privacy and security concerns in the fifth-generation (5G)-enabled vehicle networks. To verify vehicles, however, these conditional privacy-preserving authentication (CPPA) systems required a roadside unit, an expensive component of vehicular networks. Moreover, these CPPA systems incur exceptionally high communication and processing costs. This study proposes a CPPA method based on fog computing (FC), as a solution for these issues in 5G-enabled vehicle networks. In our proposed FC-CPPA method, a fog server is used to establish a set of public anonymity identities and their corresponding signature keys, which are then preloaded into each authentic vehicle. We guarantee the security of the proposed FC-CPPA method in the context of a random oracle. Our solutions are not only compliant with confidentiality and security standards, but also resistant to a variety of threats. The communication costs of the proposal are only 84 bytes, while the computation costs are 0.0031 , 2.0185 to sign and verify messages. Comparing our strategy to similar ones reveals that it saves time and money on communication and computing during the performance evaluation phase.
  • ItemOpen Access
    Fraudulent Account Detection in the Ethereum’s Network Using Various Machine Learning Techniques
    (Universiti Malaysia Pahang Publishing, 2022-07-01) Sallam, Amer; Rassem, Taha H.; Abdu, Hanadi; Abdulkareem, Haneen; Saif, Nada; Abdullah, Samia
    On the Ethereum network, users communicate with one another through a variety of different accounts. Pseudo-anonymity was enforced over the network to provide the highest level of privacy. By using accounts that engage in fraudulent activity across the network, such privacy may be exploited. Like other cryptocurrencies, Ethereum blockchain may exploited with several fraudulent activities such as Ponzi schemes, phishing, or Initial Coin Offering (ICO) exits, etc. However, the identification of parameters with abnormal account characteristics is not an easy task and requires an intelligent approach to distinguish between normal and fraudulent activities. Therefore, this paper has attempted to solve this a problem by using machine learning techniques to introduce a robust approach that can detect fraudulent accounts on Ethereum. We have used a K-Nearest Neighbor, Random Forest and XGBoost over a collected dataset of 4,681 instances along with 2,179 fraudulent accounts associated and 2,502 regular accounts. The XGBoost, RF, and KNN techniques achieved average accuracies of 96.80 %, 94.8 8%, and 87.85% and an average AUC of 0.995, 0.99 and 0.93, respectively.
  • ItemOpen Access
    Multi-Method Diagnosis of CT Images for Rapid Detection of Intracranial Hemorrhages Based on Deep and Hybrid Learning
    (MDPI, 2022-08-07) Mohammed, Badiea Abdulkarem; Senan, Ebrahim Mohammed; Al-Mekhlafi, Zeyad Ghaleb; Rassem, Taha H.; Makbol, Nasrin M.; Alanazi, Adwan Alownie; Almurayziq, Tariq S.; Ghaleb, Fuad A.; Sallam, Amer A.
    Intracranial hemorrhaging is considered a type of disease that affects the brain and is very dangerous, with high-mortality cases if there is no rapid diagnosis and prompt treatment. CT images are one of the most important methods of diagnosing intracranial hemorrhages. CT images contain huge amounts of information, requiring a lot of experience and taking a long time for proper analysis and diagnosis. Thus, artificial intelligence techniques provide an automatic mechanism for evaluating CT images to make a diagnosis with high accuracy and help radiologists make their diagnostic decisions. In this study, CT images for rapid detection of intracranial hemorrhages are diagnosed by three proposed systems with various methodologies and materials, where each system contains more than one network. The first system is proposed by three pretrained deep learning models, which are GoogLeNet, ResNet-50 and AlexNet. The second proposed system using a hybrid technology consists of two parts: the first part is the GoogLeNet, ResNet-50 and AlexNet models for extracting feature maps, while the second part is the SVM algorithm for classifying feature maps. The third proposed system uses artificial neural networks (ANNs) based on the features of the GoogLeNet, ResNet-50 and AlexNet models, whose dimensions are reduced by a principal component analysis (PCA) algorithm, and then the low-dimensional features are combined with the features of the GLCM and LBP algorithms. All the proposed systems achieved promising results in the diagnosis of CT images for the rapid detection of intracranial hemorrhages. The ANN network based on fusion of the deep feature of AlexNet with the features of GLCM and LBP reached an accuracy of 99.3%, precision of 99.36%, sensitivity of 99.5%, specificity of 99.57% and AUC of 99.84%.
  • ItemOpen Access
    Deep Learning and Machine Learning for Early Detection of Stroke and Haemorrhage
    (Tech Science Press, 2022-02-24) Al-Mekhlafi, Zeyad Ghaleb; Senan, Ebrahim Mohammed; Rassem, Taha H.; Mohammed, Badiea Abdulkarem; Makbol, Nasrin M.; Alanazi, Adwan Alownie; Almurayziq, Tariq S.; Ghaleb, Fuad A.
    Stroke and cerebral haemorrhage are the second leading causes of death in the world after ischaemic heart disease. In this work, a dataset containing medical, physiological and environmental tests for stroke was used to evaluate the efficacy of machine learning, deep learning and a hybrid technique between deep learning and machine learning on the Magnetic Resonance Imaging (MRI) dataset for cerebral haemorrhage. In the first dataset (medical records), two features, namely, diabetes and obesity, were created on the basis of the values of the corresponding features. The t-Distributed Stochastic Neighbour Embedding algorithm was applied to represent the high-dimensional dataset in a low-dimensional data space. Meanwhile,the Recursive Feature Elimination algorithm (RFE) was applied to rank the features according to priority and their correlation to the target feature and to remove the unimportant features. The features are fed into the various classification algorithms, namely, Support Vector Machine (SVM), K Nearest Neighbours (KNN), Decision Tree, Random Forest, and Multilayer Perceptron. All algorithms achieved superior results. The Random Forest algorithm achieved the best performance amongst the algorithms; it reached an overall accuracy of 99%. This algorithm classified stroke cases with Precision, Recall and F1 score of 98%, 100% and 99%, respectively. In the second dataset, the MRI image dataset was evaluated by using the AlexNet model and AlexNet + SVM hybrid technique. The hybrid model AlexNet + SVM performed is better than the AlexNet model; it reached accuracy, sensitivity, specificity and Area Under the Curve (AUC) of 99.9%, 100%, 99.80% and 99.86%, respectively.
  • ItemOpen Access
    Eye Tracking-Based Diagnosis and Early Detection of Autism Spectrum Disorder Using Machine Learning and Deep Learning Techniques
    (MDPI, 2022-02-10) Ahmed, Ibrahim Abdulrab; Senan, Ebrahim Mohammed; Rassem, Taha H.; Ali, Mohammed A. H.; Shatnawi, Hamzeh Salameh Ahmad; Alwazer, Salwa Mutahar; Alshahrani, Mohammed
    Eye tracking is a useful technique for detecting autism spectrum disorder (ASD). One of the most important aspects of good learning is the ability to have atypical visual attention. The eye-tracking technique provides useful information about children’s visual behaviour for early and accurate diagnosis. It works by scanning the paths of the eyes to extract a sequence of eye projection points on the image to analyse the behaviour of children with autism. In this study, three artificial-intelligence techniques were developed, namely, machine learning, deep learning, and a hybrid technique between them, for early diagnosis of autism. The first technique, neural networks [feedforward neural networks (FFNNs) and artificial neural networks (ANNs)], is based on feature classification extracted by a hybrid method between local binary pattern (LBP) and grey level co-occurrence matrix (GLCM) algorithms. This technique achieved a high accuracy of 99.8% for FFNNs and ANNs. The second technique used a pre-trained convolutional neural network (CNN) model, such as GoogleNet and ResNet-18, on the basis of deep feature map extraction. The GoogleNet and ResNet-18 models achieved high performances of 93.6% and 97.6%, respectively. The third technique used the hybrid method between deep learning (GoogleNet and ResNet-18) and machine learning (SVM), called GoogleNet + SVM and ResNet-18 + SVM. This technique depends on two blocks. The first block used CNN to extract deep feature maps, whilst the second block used SVM to classify the features extracted from the first block. This technique proved its high diagnostic ability, achieving accuracies of 95.5% and 94.5% for GoogleNet + SVM and ResNet-18 + SVM, respectively.
  • ItemOpen Access
    Early Diagnosis of Brain Tumour MRI Images Using Hybrid Techniques between Deep and Machine Learning
    (Wiley, 2022-05-18) Senan, Ebrahim Mohammed; Jadhav, Mukti E.; Rassem, Taha H.; Aljaloud, Abdulaziz Salamah; Mohammed, Badiea Abdulkarem; Al-Mekhlafi, Zeyad Ghaleb; Luminita Moraru
    Cancer is considered one of the most aggressive and destructive diseases that shortens the average lives of patients. Misdiagnosed brain tumours lead to false medical intervention, which reduces patients’ chance of survival. Accurate early medical diagnoses of brain tumour are an essential point for starting treatment plans that improve the survival of patients with brain tumours. Computer-aided diagnostic systems have provided consecutive successes for helping medical doctors make accurate diagnoses and have conducted positive strides in the field of deep and machine learning. Deep convolutional layers extract strong distinguishing features from the regions of interest compared with those extracted using traditional methods. In this study, different experiments are performed for brain tumour diagnosis by combining deep learning and traditional machine learning techniques. AlexNet and ResNet-18 are used with the support vector machine (SVM) algorithm for brain tumour classification and diagnosis. Brain tumour magnetic resonance imaging (MRI) images are enhanced using the average filter technique. Then, deep learning techniques are applied to extract robust and important deep features via deep convolutional layers. The process of combining deep and machine learning techniques starts, where features are extracted using deep learning techniques, namely, AlexNet and ResNet-18. These features are then classified using SoftMax and SVM. The MRI dataset contains 3,060 images divided into four classes, which are three tumours and one normal. All systems have achieved superior results. Specifically, the AlexNet+SVM hybrid technique exhibits the best performance, with 95.10% accuracy, 95.25% sensitivity, and 98.50% specificity.
  • Item
    A novel inertia moment estimation algorithm collaborated with Active Force Control scheme for wheeled mobile robot control in constrained environments
    (Elsevier, 2021-06-21) Ali, Mohammed A.H.; Radzak, Muhammad S.A.; Mailah, Musa; Yusoff, Nukman; Razak, Bushroa Abd; Karim, Mohd Sayuti Ab.; Ameen, Wadea; Jabbar, Waheb A.; Alsewari, AbdulRahman A.; Rassem, Taha H.; Nasser, Abdullah B.; Abdulghafor, Rawad
    This paper presents a novel inertia moment estimation algorithm to enable the Active Force Control Scheme for tracking a wheeled mobile robot (WMR) effectively in a specific trajectory within constrained environments such as on roads or in factories. This algorithm, also known as laser simulator logic, has the capability to estimate the inertia moment of the AFC-controller when the robot is moving in a pre-planned path with the presence of noisy measurements. The estimation is accomplished by calculating the membership function based on the experts’ views in any form (symmetric or non-symmetric) with lowly or highly overlapped linguistic variables. A new Proportional-Derivative Active Force Controller (PD-AFC-LS-QC), employing the use of laser simulator logic and quick compensation loop, has been developed in this paper to robustly reject the noise and disturbances. This controller has three feedback control loops, namely, internal, external and quick compensation loops to compensate effectively the disturbances in the constrained environments. A simulation and experimental studies on WMR path control in two kinds of environments; namely, zigzag and highly curved terrains, were conducted to verify the proposed algorithm and controller which was then compared with other existed control schemes. The results of the simulation and experimental works show the capability of the proposed algorithms and the controller to robustly move the WMR in the constrained environments.
  • ItemEmbargo
    A New Wavelet Completed Local Ternary Count (WCLTC) for Image Classification
    (IEEE, 2021-11-01) Rassem, Taha H.; Alkareem, Fatimah A.; Mohammed, Mohammed Falah; Makbol, Nasrin M.; Sallam, Amer
    To overcome noise sensitivity and increase the discriminative quality of the Local Binary Pattern, a Completed Local Ternary Count (CLTC) was developed by combining the Local Ternary Pattern (LTP) with the Completed Local Binary Count (CLBC) (LBP). Furthermore, by integrating the proposed CLTC with the Redundant Discrete Wavelet Transform (RDWT) to construct a Wavelet Completed Local Ternary Count, the proposed CLTC’s discriminative property is improved (WCLTC). As a result, more accurate local texture feature capture inside the RDWT domain is possible. The proposed WCLTC is utilised to perform texture and medical image classification tasks. The WCLTC performance is evaluated using two benchmark texture datasets, CUReT and Outex, as well as three medical picture databases, 2D Hela, VIRUS Texture, and BR datasets. With several of these datasets, the experimental findings demonstrate a remarkable classification accuracy. In several cases, the WCLTC performance outperformed the prior descriptions. With the 2D Hela, CUReT, and Virus datasets, the WCLTC achieves the highest classification accuracy of 96.91%, 97.04 percent, and 72.89%, respectively.
  • ItemOpen Access
    Multi-Method Analysis of Medical Records and MRI Images for Early Diagnosis of Dementia and Alzheimer’s Disease Based on Deep Learning and Hybrid Methods
    (MDPI, 2021-11-20) Mohammed, Badiea Abdulkarem; Senan, Ebrahim Mohammed; Rassem, Taha H.; Makbol, Nasrin M.; Alanazi, Adwan Alownie; Al-Mekhlafi, Zeyad Ghaleb; Almurayziq, Tariq S.; Ghaleb, Fuad A.
    Dementia and Alzheimer’s disease are caused by neurodegeneration and poor communication between neurons in the brain. So far, no effective medications have been discovered for dementia and Alzheimer’s disease. Thus, early diagnosis is necessary to avoid the development of these diseases. In this study, efficient machine learning algorithms were assessed to evaluate the Open Access Series of Imaging Studies (OASIS) dataset for dementia diagnosis. Two CNN models (AlexNet and ResNet-50) and hybrid techniques between deep learning and machine learning (AlexNet+SVM and ResNet-50+SVM) were also evaluated for the diagnosis of Alzheimer’s disease. For the OASIS dataset, we balanced the dataset, replaced the missing values, and applied the t-Distributed Stochastic Neighbour Embedding algorithm (t-SNE) to represent the high-dimensional data in the low-dimensional space. All of the machine learning algorithms, namely, Support Vector Machine (SVM), Decision Tree, Random Forest and K Nearest Neighbours (KNN), achieved high performance for diagnosing dementia. The random forest algorithm achieved an overall accuracy of 94% and precision, recall and F1 scores of 93%, 98% and 96%, respectively. The second dataset, the MRI image dataset, was evaluated by AlexNet and ResNet-50 models and AlexNet+SVM and ResNet-50+SVM hybrid techniques. All models achieved high performance, but the performance of the hybrid methods between deep learning and machine learning was better than that of the deep learning models. The AlexNet+SVM hybrid model achieved accuracy, sensitivity, specificity and AUC scores of 94.8%, 93%, 97.75% and 99.70%, respectively.
  • ItemEmbargo
    Iris Recognition System Using Convolutional Neural Network
    (IEEE, 2021-08-24) Sallam, Amer; Al Amery, Hadeel; Al-Qudasi, Somaia; Al-Ghorbani, Safaa; Rassem, Taha H.; Makbol, Nasrin M.
    Identification system is one of the important parts in security domains of the present time. The traditional protection methods considered to be inefficient and unreliable as they are subjected to the theft, imitation or forgetfulness. In contrast, biometrics such as facial recognition, fingerprints and the retina have emerged as modern protection methods, but still also suffer from some defects and violations. However, Iris recognition is an automated method that considered as a promising biometric identification due to the stability and the uniqueness of its patterns. In this paper, an iris recognition system based on Convolutional Neural Network (CNN) model was proposed. CNN is used to perform the required processes of feature extraction and classification. The proposed system was evaluated through CASIA-V1 and ATVS datasets, after the required pre-processing steps taken place, and achieved 98% and 97.83% as a result, respectively.
  • ItemEmbargo
    Reliability-driven large group consensus decision-making method with hesitant fuzzy linguistic information for the selection of hydrogen storage technology
    (Elsevier, 2024-09-12) Wang, Peng; Dong, Xin; Chen, Junhong; Wu, Xiaoming; Chiclana, Francisco
    The use of hydrogen storage technology (HST) as a bridge for producing and utilizing hydrogen energy in the hydrogen industry chain is significant, and its evaluation has attracted the interest of researchers. Since there are different types of HSTs, selecting the most appropriate one requires the participation of plenty of experts with different professional backgrounds, which makes this be modeled as a large group decision-making problem. This paper develops a reliability-driven large group consensus decision-making (LGCDM) method for HST selection using the hesitant fuzzy linguistic terms set (HFLTS) as the evaluation representation format. Specifically, the expertise level of individuals and the reliability of group opinions are measured based on the set variables, and then the dimensionality of large groups is reduced based on the reliability of subgroup opinions. Furthermore, an opinion reliability rating mechanism is designed and, when consensus is not satisfactory, a feedback recommendation mechanism and consensus optimization mechanism are developed for implementation. Finally, the proposed reliability-driven LGCDM approach is applied to the HST selection for THVOW Company, and the comparison with existent related approaches indicates that it not only is practical and reasonable, but also provides a technical path for relevant departments to make decisions on practical issues.
  • ItemEmbargo
    Minimum Cost Consensus-Based Social Network Group Decision Making With Altruism-Fairness Preferences and Ordered Trust Propagation
    (IEEE, 2024-09-17) Feng, Yu; Dang, Yaoguo; Wang, Junjie; Du, Junliang; Chiclana, Francisco
    Different from conventional decision-making environments, decision makers (DMs) in a community setting usually exhibit the complex social preferences and intricate social interactions, which may lead to high-decision costs for group consensus reaching. To address this challenge, we design a minimum cost consensus-based social network group decision making (SNGDM) approach considering altruism-fairness preferences and ordered trust propagation. First, a trust propagation method with order effect and path length is proposed to estimate the completed trust relationships among DMs in order to determine the weights of DMs. Then, inspired by the interaction of altruism and fairness preferences, we define the individual altruism-fairness preference utility function and utility level for cost consensus, and explore some properties. Afterwards, a new minimum cost consensus-based SNGDM with individual altruism-fairness preference utility is constructed. Finally, the validity of the proposed consensus framework is confirmed through the carbon reduction consensus problem of China’s aviation enterprises. Moreover, the sensitivity studies and comparative analysis are conducted to further demonstrate the merits of our proposal.