Forthcoming and Online First Articles
International Journal of Information and Communication Technology
Forthcoming articles have been peer-reviewed and accepted for publication but are pending final changes, are not yet published and may not appear here in their final order of publication until they are assigned to issues. Therefore, the content conforms to our standards but the presentation (e.g. typesetting and proof-reading) is not necessarily up to the Inderscience standard. Additionally, titles, authors, abstracts and keywords may change before publication. Articles will not be published until the final proofs are validated by their authors.
Forthcoming articles must be purchased for the purposes of research, teaching and private study only. These articles can be cited using the expression "in press". For example: Smith, J. (in press). Article Title. Journal Title.
Articles marked with this shopping trolley icon are available for purchase - click on the icon to send an email request to purchase.
Online First articles are published online here, before they appear in a journal issue. Online First articles are fully citeable, complete with a DOI. They can be cited, read, and downloaded. Online First articles are published as Open Access (OA) articles to make the latest research available as early as possible.
Articles marked with this Open Access icon are Online First articles. They are freely available and openly accessible to all without any restriction except the ones stated in their respective CC licenses.Register for our alerting service, which notifies you by email when new issues are published online.
International Journal of Information and Communication Technology (20 papers in press) Regular Issues
Abstract: With the continuous progress of social economy, the pursuit and demand of people for art are getting higher and higher. However, due to the limitations of designers understanding, technical skill levels, and costs, they are unable to meet all the demands of users. In view of these insufficiencies and limitations, the technology related to virtual reality (VR) and big data image processing is introduced in this paper. Our innovation lies in the integration of these technologies to establish a digital model of space based on geographical location, which not only consolidates the business logic in the visual expression of multimedia art patterns but also implements the integration of indoor and outdoor spatial information drawing. By creating virtual scenes, we implement 3D image reconstruction of complex scenes, achieving real scene experience and obtaining better visual communication effects. The uniqueness of our approach is further demonstrated through simulation experiments, indicating that the proposed virtual reality technology and big data image processing are highly effective and can support computer multimedia art patterns and visual communication design. Keywords: virtual reality; VR; visual communication; image processing; landscape reconstruction.
Abstract: This paper models the monitoring and early warning task of surface subsidence in the high-voltage line tower area, this study uses PS-InSAR technology to preprocess the clipped satellite remote sensing images to obtain surface subsidence datathe predicted values and deformation curves are basically coincident, and the trend of change is basically consistent. The absolute error of settlement prediction is less than 1%, indicating high prediction accuracy. From the experiment, it can be seen that this improvement effectively improves the system accuracy and reduces prediction errors. From the experimental results, it can be seen that the monitoring and early warning model of regional subsidence of transmission line based on the time series algorithm proposed in this paper has a certain effect, and can meet the needs of monitoring and early warning of regional subsidence of transmission lines. Keywords: time series; transmission lines; regional subsidence; monitoring; early warning.
Abstract: Icing will not only increase the weight of the conductor to make the conductor bear greater vertical gravity and make the axial tension greater when the conductor gallops, but also increase the cross-sectional area of the conductor to increase the aerodynamic force of the conductor. Starting from the basic principles of distributed optical fibres and measurable physical quantities, this paper uses the method of combining theoretical analysis and simulation modelling to establish the mathematical and physical models and signal characteristic models of the multi-dimensional physical quantities (temperature, strain, vibration and light polarisation state) of distributed optical fibre monitoring and line icing and wind vibration events. Moreover, this paper constructs a multi-parameter distributed optical fibre sensing system according to the monitoring requirements, which solves the problem of low signal-to-noise ratio in long-distance monitoring. Finally, through experimental verification, it is found that the model proposed in this paper has a good intelligent monitoring effect on the icing status of transmission lines, which can provide a reference for real-time monitoring of transmission lines. Keywords: vibration model; transmission lines; icing; intelligent monitoring.
Abstract: This article proposes a method for monitoring the mental health status of college students based on machine learning models. By integrating multidimensional data such as psychological assessment questionnaires and daily behavior data, and using machine learning techniques such as support vector machine, random forests, and deep learning algorithms, a prediction model that can efficiently identify the mental health status of college students is constructed. This model improves the quality of data and the accuracy of model predictions through steps such as feature engineering and data preprocessing. This article used real datasets from multiple universities for experimental testing, and the results showed that the method performed well in multiple evaluation indicators such as accuracy, recall, and F1 score, demonstrating strong practicality and promotional value. Keywords: deep learning; machine learning; mental health; college student psychology.
Abstract: This article proposes a digital dance generation method based on mixture density network (MDN), aiming to effectively capture and generate complex dance action sequences. Firstly, we analysed the temporal dependencies and diverse features of dance movements, and designed a multimodal temporal generation framework using MDN and long short-term memory (LSTM) networks to capture dynamic correlations and pose changes between dance movements. This framework can generate action sequences that match the music style when inputting music or rhythm information, with high continuity, coordination, and naturalness. This paper assesses the generated dance motions by the model using a publicly available dance dataset, and verified the effectiveness of this method through subjective and objective quantitative indicators. The experimental results show that compared to traditional generative models, the MDN based model has improved the fluency, naturalness, and diversity of generated actions. Keywords: deep learning; computer music choreography; feature extraction; action filtering.
Abstract: As information technology develops, the use of mobile learning tools in English instruction is spreading rather rapidly. This study combines data mining techniques to explore effective methods for optimising English learning modes on mobile learning platforms. Firstly, using K-means clustering analysis to identify students learning behaviour characteristics and identify student groups with different learning styles; Then, the Apriori algorithm is applied to mine the association rules between learning behaviour and learning outcomes, further revealing the key factors that affect learning outcomes. In addition, tracking students learning progress and emotional changes through time series analysis provides a design basis for personalised learning paths for the platform. The research results indicate that optimisation strategies based on data mining can help improve learning effectiveness and support teachers and platform developers to provide more targeted learning suggestions. Keywords: K-Means; Apriori; data mining; time series analysis.
Abstract: The difficulty of unbalanced datasets in classification issues has become more noticeable with the fast expansion of data science and machine learning approaches. When confronted with uneven data, conventional machine learning methods often produce poor prediction of a few classes. Based on Bayesian optimisation (BO), we propose in this work an enhanced convolutional neural network (CNN) framework (BO-CNN) meant to optimise the hyperparameter configuration of CNNs while resolving the class bias problem in unbalanced data. Experimental results reveal that BO-CNN shows benefits on challenging datasets, lowers miss-detection and false alarms, and efficiently enhances the capacity of the model to manage unbalanced data. These results offer a fresh approach for unbalanced data categorisation and a useful guide for the future optimisation and implementation of deep learning models. Keywords: BO; Bayesian optimisation; convolutional neural network; CNN; unbalanced data recognition.
Abstract: Personalised book recommendations have become a major difficulty given the fast growth of digital resources and online libraries. This work presents a hybrid recommender system based on extreme learning machine (ELM) to improve the accuracy and diversity of book recommendations in libraries. Combining content-based filtering and collaborative filtering, the system uses the advantages of both techniques above their respective restrictions. The suggested system effectively handles extensive user behaviour and book information by including ELM, which provides fast training and high generalising capacity. Comparatively to conventional approaches, experimental data reveal that the hybrid model considerably increases suggestion accuracy, diversity, and coverage. Key parameters used in evaluation of the system include precision, recall, variety, and coverage, therefore proving its possible use in library book recommendation systems. Keywords: extreme learning machine; ELM; hybrid recommender system; library books; personalised book recommendations.
Abstract: This paper aims to integrate machine vision into intelligent robots to realise the automatic separation of insulated optical units of OPGW optical cables. This paper combines improved machine vision to construct a three-dimensional space model of OPGW optical cable, combines three-dimensional measurement technology to identify the insulated optical units from the optical cable structure. The experimental results show that the operation effect of the system is very remarkable, the three-dimensional coordinate guided mechanical finger point accuracy measured by this binocular system reached 0.4 mm, and the maximum radial and axial loads of the tool were within therated load, which indirectly demonstrates the feasibility of this method and structural design. The robot system designed in this article reduces the damage to the metal layer of cables caused by traditional methods, lowers the risk of cable stripping. It also provides automation reference for more complexcable operations in the future. Keywords: machine vision; insulated optical unit of optical cable; automatic separation; robot.
Abstract: Convolution neural network (CNN) is a deep learning model that is widely used in image recognition, image classification. However, traditional deep learning models require extensive annotation information during training, leading to prolonged training times that directly impacts efficiency and performance. In order to solve this problem, this paper proposes a five-layer convolution neural network structure based on transfer learning to extract, train and fuse features of source images. First, an improved VGG-19 network is used to extract the preliminary features of the source images, and the training samples are transferred to the encoder for deep feature extraction by setting the VGG-19 network parameters. Then, the extracted feature samples and a five-layer U-Net neural network construct the decoder for feature reconstruction. Batch normalisation is applied to prevent over-fitting of the model. Finally, the loss function is applied layer-by-layer in supervised learning to obtain the quadratic decision graphs that are used to fuse the source images to generate the output images. The proposed model in this paper demonstrates a significant enhancement in the visual effect of images compared with other models. Keywords: deep learning; convolutional neural network; transfer learning; image fusion.
Abstract: Insulators are crucial components for the safety and reliability of power systems. However, due to their small size and complex structure, precise segmentation of insulators is challenging. To address this issue, this paper proposes a directional attention-based PointNet++ model (PDA). The core module of PDA is the directional attention (DA) module, which consists of spatial self-attention (SSA) and channel self-attention (CSA). This module is designed to establish long-range relationships in both spatial and channel directions of the feature map, enabling global modelling. Additionally, to reduce computational costs, multi-scale pyramid pooling is embedded in both the SSA and CSA modules. Notably, by integrating DA into PointNet++, the model enhances the correlation between point cloud features and the long-range dependency of positional information without significantly increasing the computational burden. Experimental results demonstrate that the PDA model significantly outperforms existing models in segmenting insulator point clouds from multiple power transmission corridors. Keywords: PointNet++; PDA; spatial self-attention; SSA; channel self-attention; CSA; insulators.
Abstract: Radar emitter recognition is a critical part of electronic countermeasures and determines the implementation of subsequent jamming measures. With the rapid development of deep learning (DL), the radar emitter recognition method based on DL is also widely developed. However, methods based on time-frequency analysis to obtain image features suffered from information loss and computational complexity; end-to-end methods based on raw radar signals had low accuracy at low signal-to-noise ratio (SNR). Therefore, we investigated the frequency distribution of signal and noise, analysed the working principle of batch norm, and proposed to suppress the high-frequency noise by removing the batch norm in the network. Furthermore, we constructed a straightforward end-to-end denoising and recognition network as well as utilised the latest classification network training process to improve the accuracy of radar emitter recognition at low SNR. Experiments validated that the proposed method achieved SOTA result on the well-known DeepSig RadioML 2018.01A. Keywords: radar emitter recognition; signal denoising; deep learning; end-to-end; time series.
Abstract: To address the domain distribution mismatch between synthetic scene text data and real-world scene text data in arbitrary orientations, we introduce SynthBendText3D - a framework based on a 3D graphics engine that synthesises scene text data in various orientations. The framework generates a large number of text instances in arbitrary directions and constructs a 3D scene to position these instances. By leveraging domain randomisation techniques, it randomises scene parameters such as object arrangement, materials, lighting, and camera angles, ensuring a high degree of diversity in the synthesised data. Moreover, the framework incorporates a polygon reconstruction algorithm to annotate each synthesised text instance with polygonal bounding boxes. Experimental results demonstrate the effectiveness of the data generated by our framework. Keywords: scene text detection; synthetic data; domain randomisation; domain adaption.
Abstract: We propose a novel deep learning-based model for predicting the remaining life of lithium-ion batteries. Existing methods merely model the remaining life's temporal changes, overlooking inherent time series periodicity and compromising prediction accuracy. Our model capitalises on multi-cycle features in time series analysis, using well-designed 2D temporal blocks to handle uncertainties in battery remaining useful life changes. It extracts complex patterns within charge and discharge cycles, achieving high-precision predictions of future battery states. On multiple common battery datasets, it surpasses existing methods in accuracy and robustness, validating its effectiveness. Keywords: lithium-ion battery; remaining useful life prediction; multi-cycle time series.
Abstract: With its natural benefits of decentralisation and immutability, blockchain technology has become rather popular in smart contracts, supply chains, and banking. It struggles greatly, nevertheless, in terms of scalability and transaction processing efficiency. Thus, in this regard, these studies present the execute-order-re-execute and validate (EOR) architecture to handle these problems. By means of off-chain execution, on-chain ordering, on-chain re-execution and verification phases, the EOR architecture maximises transaction processing, hence improving system performance and security. For nested contract concurrency it uses a two-phase locking technique, and for effective verification a lockchain architecture. Offering a significant means for extending blockchain uses, experimental results show a 40% boost in transaction processing efficiency, a 2.5% transaction abort rate, and enhanced system stability in high-conflict settings. Keywords: blockchain; concurrency control; nested contracts; on-chain-off-chain.
Abstract: Accurate and effective facial expression recognition (FER) is of great significance in fields such as intelligent monitoring and emotional computing today. This article proposes a deep learning method based on You Only Look Once Version 8 (YOLOv8), which combines YOLOv8's real-time and efficient object detection capabilities with the feature extraction advantages of convolutional neural networks (CNNs) to improve facial expression recognition performance in complex environments. Firstly, YOLOv8 is used for precise facial detection. Then, the detected facial regions are fed into a feature extraction network, which extracts high-level features related to facial expressions through deep CNN, enhancing the robustness of the model to different complex scenes. The results of the experiment indicate that our approach performs well on publicly available facial expression datasets, especially in complex scenes where it significantly outperforms traditional expression recognition methods. This model provides new ideas for future applications in diverse, dynamic, and complex environments. Keywords: facial recognition; facial expression recognition; FER; convolutional neural network; CNN; YOLOv8.
Abstract: With the rapid development of globalisation and e-commerce, cross-border e-commerce platforms are facing the challenge of improving user experience and recommendation system efficiency while meeting the diverse needs of consumers. Traditional recommendation systems rely heavily on users' historical behaviour and simple rating data. However, these methods often face problems such as data sparsity and single recommendation results in practical applications. Therefore, this article proposes a hybrid collaborative recommendation method for cross-border e-commerce products based on multidimensional evaluation, which fully utilises users' multidimensional evaluation information of products to address the complexity of cross-border e-commerce. Then, the system framework and algorithm flow were presented. Finally, the improved algorithm proposed in this paper was experimentally analysed using a cross-border e-commerce enterprise order dataset. Compared with traditional collaborative filtering recommendation algorithms, it reduced the impact of data sparsity in collaborative filtering recommendation algorithms and verified that the improved algorithm has better recommendation performance. Keywords: recommendation system; cross border e-commerce; collaborative filtering; multidimensional evaluation.
Abstract: Texts generated by college students through social sharing are characterised by emotional richness and psychological vulnerability. To address the issue that existing social-psycho-emotional profiling methods for college students rely on the size of the labelled dataset and have unsatisfactory classification results, this article first optimises semi-supervised learning (MSASL), which composes the samples and incorporates a smoothness loss while imposing consistency constraints. The BERT model is then used to obtain a semantic representation of the social text, using interactive attention to capture important feature information related to the opinion tendency of the topic words. Finally, semi-supervised GAN (MSASL-GAN) is applied to optimise the text feature representation, and the sentiment feature classification results are output through the fully connected layer. The experimental results show that the classification accuracy of the proposed method is improved by 5.01%-11.51% compared with the comparison model. Keywords: sentiment profiling; semi-supervised learning; SSL; BERT model; interactive attention; generative adversarial network; GAN. Interactive decision support system with machine intelligence for augmentative communication by Ruiwei Chen, C.B. Sivaparthipan Abstract: Many augmentative communication technologies help physically challenged people to communicate with others in the present world. Augmentative communication system integrates components that include symbols, strategies, and aids that enhance communication abilities. Augmentative communication technology’s significant challenge for physically challenged people is the lack of speech expression and depression. Interactive decision support system integrated machine intelligence framework (DSS-MIF) supports augmentative communication proposed to express the physically challenged expression and depression. DSS has multiple sensors, which monitor the heartbeat rate, vocal cord vibration, body temperature, and muscle contraction. The related data are calibrated using MIF, in which the expression of the person is recognised. Based on the DSS-MIF output, physically challenged people could express themselves to others using the augmentative communication system. The experimental analysis shows that the proposed DDS-MIF for augmentative communication improves performance rate to 98.66% and shows physically challenged people’s expression effectively. Keywords: augmentative communication; machine intelligence; decision support; multiple sensors. DOI: 10.1504/IJICT.2023.10056637 A new fast DBSCAN using dual-space analysis and colour integral volume for document image segmentation by Zakia Kezzoula, Djamel Gaceb Abstract: The segmentation of the colour document images is an essential step allowing facilitating and improving the stages of characterisation and interpretation of the information contained in these documents. Recent systems of automatic processing and recognition of document images, which use separation of colouremric layers, are more efficient compared to conventional systems, only based on binary or grey levels images. This task requires non-supervised pixel segmentation or clustering techniques to separate the document image to a variable and unknown number of colour layers. The methods based on density are widely used in this context at pixel level, such as the DBSCAN method and its different variants, very robust to the noise and more adapted to the degradations present on document images, but who suffer from a great complexity. In this context, we propose a new faster DBSCAN variant using the volume integral in colourimitric space for the first time to significantly reduce calculation time. The combination of the two spaces, Cartesian and colorimetric has also accelerated the method and improved its performance on document images with different challenges. The results obtained show the effectiveness of the proposed approach, which was marked by significant improvement in the quality of segmentation and reduction in computed time. Keywords: clustering; DBSCAN; region growing; document image segmentation; fast I2SDBSCAN; 3D colour histogram; integral volume. DOI: 10.1504/IJICT.2024.10065387 |
Open Access
Keep up-to-date
-
Our Blog -
Follow us on Twitter -
Visit us on Facebook - Our Newsletter (subscribe for free)
- RSS Feeds
- New issue alerts
- Inderscience is a member of publishing organisations including:
-
-
