International Journal of Computer Applications in Technology (34 papers in press)

Regular Issues

• Open Access

  Digital media video image data processing based on computer vision ( Free Full-text Access )

  by Xiawei Lu 
  Abstract: This paper describes the application direction of DMVI processing technology and the acquisition and post-processing of ultra-high-definition quality data, explores the application of DMVI processing technology in image analysis, proposes a method for obtaining ultra-high definition quality video data, and discusses the reconstruction of ultra-high-definition quality video. According to the research results, satisfaction with the introduction of the five-dimensional light field function algorithm and CV technology reached over 21%; at 4K resolution, the processing time of the five-dimensional light field was 1.05
  Keywords: digital media video image; image data processing; computer vision; ultra clear picture quality image.
  DOI: 10.1504/IJCAT.2025.10073493
   
  
• FPGA implementation and Multisim simulation of a new four-dimensional two-scroll hyperchaotic system with coexisting attractors  
  by Sundarapandian Vaidyanathan, Esteban Tlelo-Cuautle, Khaled Benkouider, Aceng Sambas, Ciro Fabian Bermudez-Marquez, Samy Abdelwahab Safaan 
  Abstract: Field-programmable gate array (FPGA) design of a new four-dimensional two-scroll hyperchaotic system is investigated in this work. A detailed system modelling of the new system with a hyperchaotic attractor begins this work with phase plots, which is followed by a bifurcation study of the new system. Special dynamic properties such as multistability and symmetry are also investigated for the new system. Using Multisim software, a circuit model is designed and simulated for the new hyperchaotic system. FPGA design and Multisim simulation of the new system enable practical applications in science and engineering. The implementation of the FPGA design in this work is carried out by applying two numerical schemes, viz. Forward Euler and Trapezoidal methods. Experimental attractors observed in the oscilloscope show good match with the Matlab signal plots.The FPGA hardware resources are detailed for both numerical methods.
  Keywords: hyperchaos; bifurcation; symmetry; phase plots; hyperchaotic system;rnparameters; stability; multistability; circuit model; FPGA implementation.
  
  
• Improving hybrid-layer convolutional neural network system for lung cancer nodule classification using enhanced weight optimisation algorithm  
  by Vikul Pawar, P. Premchand 
  Abstract: In recent times, lung cancer is evolving as a highly life-threatening disease for human beings. According to the WHO, lung cancer disease is the second largest cause of deaths as compared to all other types of cancer. The prevailing available technology is striving to get more exposure in the field of medical science using Computer Assisted Diagnosis (CAD), where image processing is playing a crucial role for detecting the cancerous nodules in computer tomographic images. Augmenting the machine learning techniques with image processing algorithms is becoming a more comprehensive examination of cancer disease in proposed CAD systems. This paper is describes a heuristic approach for lung cancer nodule detection, and the proposed model predominantly consists of the following tasks, which are image enhancement, segmenting ROI (Region of Interest), features extraction, and nodule classification. In pre-processing, primarily the Adaptive Median Filter (AMF) filtering method is applied to eliminate the speckle noise from input CT images of Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): in the LIDC-IDRI dataset, the quality of input image is improved by applying Histogram Equalization (HE) technique with Contrast-Limited Adaptive (CLA) approach. Secondly, in the successive stage the Improved Level-Set (ILS) algorithm is used to segment the ROI. Furthermore, the third step of the projected work is applied to extract the definite learnable texture features and statistical features from the segmented ROI. The extracted features in the subsequent stage of classification are applied to Hybrid-Layer Convolutional Neural Network (HL-CNN) architecture to classify the lung cancer nodule as either benign or malignant. Principally this research is carried out by contributing to each stage of it, where the novel concept of the improved Hybrid-Layer Convolutional Neural Network (HL-CNN) is employed by optimising and selecting the optimal weight using the Enhanced Cat Swarm Optimisation (ECSO) algorithm. The experimental result of the proposed HL-CNN using the weight optimisation algorithm ECSO is achieved an accuracy of 93%, which is comparatively efficient with respect to existing models such as DBN, SVM, CNN, WOA, MFO, and CSO. Moreover, the proposed model conclusively gives a decision on the detected nodule as either benign or malignant.
  Keywords: Computer Assisted Diagnosis (CAD); Computer Vision; Cancer Diagnosis; Image Classification; Image Enhancement; Image Segmentation; Feature Extraction.
  
  
• Prediction model for total amount of coke oven gas generation based on FCM-RBF  
  by Lili Feng, Jun Peng, Zhaojun Huang 
  Abstract: The rational use of Coke Oven Gas (COG) is of great significance to improve the economic efficiency of enterprises. In this paper, a COG generation prediction model based on fuzzy C-mean clustering (FCM) and radial basis function (RBF) neural network is proposed to address the problems such as the difficulty of accurate modelling of COG generation process and the difficulty of real-time flow prediction. Firstly, the coke oven production process is analysed and correlation analysis is used to select the influencing factors. Secondly, the FCM is used to classify the working conditions of the coke oven, and the appropriate number of working conditions is selected through experiments. Finally, the prediction models under different working conditions are established separately by using RBF. The experiments were carried out using actual industrial production data, and the experimental results showed that the model could provide guidance reference for the dispatchers.
  Keywords: coking oven process; fuzzy C-means clustering; prediction model; radial basis function neural network.
  
  
• Advancing financial security: integrating AI and blockchain for cloud network protection in supply chain financing  
  by Weishuang Xu, Daming Li 
  Abstract: This paper investigates the integration of Artificial Intelligence (AI) into cloud network security protocols, recognizing its capacity to analyze vast datasets and detect patterns effectively. By leveraging AI, cloud network security can be fortified, thereby enhancing risk identification and control within enterprise supply chain financing, a domain where efficient data collection and integration remain pivotal challenges. The study explores the application of blockchain technology (BT) in enterprise supply chain systems, demonstrating its potential to enable real-time monitoring and management of the financing process. The research findings indicate substantial improvements in risk identification (0.99) and security performance (1.22), affirming the feasibility of integrating IoT-based BT into enterprise supply chain financing for enhanced risk management and security.
  Keywords: corporate supply chain; risk identification and control; internet of things; blockchain; electricity system trading.
  DOI: 10.1504/IJCAT.2025.10069456
   
  
• Development of intelligent data collection and management system based on internet of things big data crawler technology  
  by Jin Chen, Yao Li, Xia Hua, Long Lu 
  Abstract: This paper used web crawler technology to develop intelligent data collection and management system. This paper first analyzed the basic principles of system design and the structural requirements of the system according to the system requirements, and then evaluated the overall results of the intelligent data collection and management system. The process and principle of information collection using the IoT Big Data (BD) crawler technology were introduced in detail. Finally, the information collection effect of the system was verified by experiments. The experimental results showed that the data acquisition accuracy of the system was high, accounting for more than 90%. The system had high usability and efficiency; users were satisfied with the system, and data collection and management could be carried out well.
  Keywords: intelligent data collection; data management; big data of the internet of things; big data crawler technology.
  DOI: 10.1504/IJCAT.2025.10069544
   
  
• Visual evaluation of tourism ecological and environmental protection green management in the context of sustainable development  
  by Kun Zheng, Lijing Zhang 
  Abstract: The tourism industry has developed well in the past few years. However, due to the further spread of the COVID-19 in recent years, the development of tourism industry in various regions is relatively poor. At the same time, some behaviors of tourists in tourism have also brought a greater burden to the ecological environment of various regions. This is mainly due to the differences of the ecological environment between different regions, which leads to some habitual or other uncivilized behaviors that would cause a destructive blow to the local ecological environment. However, with the increasing scale of the tourism industry, people’s economic level and material conditions are getting better and better, and more and more people want to visit the cultural scenery of different regions, which leads to the increasingly obvious impact of the tourism industry on the ecological environment of different regions.
  Keywords: tourism ecology; green management; sustainable development; visualisation analysis.
  DOI: 10.1504/IJCAT.2025.10069655
   
  
• A lightweight authentication architecture for local communication of power terminal based on quantum key  
  by Chao Chen, Jin Qian, Shaojie Luo 
  Abstract: In recent years, there has been a notable acceleration in the construction of new power systems, particularly those incorporating renewable energy sources. At present, the scheme to bolster the secure communication capabilities of power terminals through a fusion of quantum cryptography and lightweight identification key technology is grappling with practical deployment challenges. To address this predicament, this study introduces a novel lightweight authentication architecture for power terminals within local communication networks, anchored on quantum key principles. By leveraging a secure encryption chip in conjunction with lightweight authentication key technology, the proposed architecture facilitates efficient and secure connectivity among a multitude of terminals in local communication settings. A comprehensive security analysis of the proposed solution is also presented.
  Keywords: local communication; quantum cryptography; secure chip; lightweight authentication key technology; authentication.
  DOI: 10.1504/IJCAT.2025.10070796
   
  
• Construction of intelligent tourism public service platform featuring communication big data and internet of things  
  by Shuo Liang, Yiran Wang, Lili Liu 
  Abstract: Tourism has evolved into a modern service industry and strategic pillar, driven by advancements in information and communication technology. This transformation has ushered in the era of intelligent tourism. The study introduces the background of intelligent tourism, summarising academic research on tourism public service platforms, big data (BD), and the Internet of Things (IoT). An algorithm model is proposed to establish a theoretical framework for constructing an intelligent tourism platform characterised by BD and IoT. Factor analysis identifies key elements for platform development, followed by simulation experiments to validate its effectiveness. Results demonstrate that the proposed platform increases user satisfaction by 20% compared to traditional systems. In the BD era, with growing socioeconomic and technological progress, integrating information technology across industries has become critical. Developing intelligent tourism platforms represents a significant theoretical and practical challenge for the tourism industrys advancement.
  Keywords: intelligent tourism; big data; internet of things; communication technology; wireless communication.
  DOI: 10.1504/IJCAT.2025.10071195
   
  
• Transforming design elements and enhancing visual communication via machine learning-based graphic design  
  by Zhenyuan Hao, Xiaoyang Shen 
  Abstract: This paper proposes convolutional efficient transformer-based image feature extraction. First, it combines convolution from different angles to introduce translation invariance, scale invariance, and locality. Then, a lightweight convolution module with depthwise convolution and atrous convolution is used to change the traditional processing of the input image of the transformer to accelerate the convergence speed and improve the stability. Aiming at the problem that cycle generative adversarial network (CycleGAN) has low texture clarity when processing images, the convolutional efficient transformer-based image feature extraction algorithm is added to the generator of GAN to enhance the effect of CycleGAN in image style transfer. Results demonstrate that the proposed algorithm performs well regarding top-1 accuracy, number of parameters, floating-point operations, and convergence rate. It can effectively analyze and classify visual elements, providing valuable insights and aiding in tasks such as image categorization, style transfer, generative design, or automated design recommendation systems.
  Keywords: machine learning; graphic design; transformer; CycleGAN; Image style transfer.
  DOI: 10.1504/IJCAT.2025.10071201
   
  
• RNN-PSO: a tuned neural network with optimisation algorithm for keratoconus classification  
  by G.P. Ramesh, Priyanka Subramanian, B. Ramakantha Reddy 
  Abstract: Keratoconus is the medical illness wherein the cornea attains a conical shape caused by thinning of corneal stroma. Symptoms based on keratoconus stage gets unnoticed. The advanced phases are found through vision loss and protrusion. The diagnosis of Keratoconus and its severity classification from corneal topography images has gained importance with advanced imaging and machine learning. Convolutional Neural Networks (CNN) is used for extracting significant features from images categorize the keratoconus. Then, extracted features are classified using Recurrent Neural Network (RNN) and the hyper parameters of the RNN is optimized using Particle Swarm Optimization (PSO) which improves the keratoconus classification performance. The performance of proposed method is evaluated with four classes namely normal, subclinical, keratoconus and advanced keratoconus in terms of accuracy, sensitivity and specificity. The method attained the accuracy of 0.80 on normal, 0.81 on subclinical, 0.85 on keratoconus and 0.91 on advanced keratoconus.
  Keywords: convolutional neural networks; keratoconus; particle swarm optimisation recurrent neural network; sub-clinical.
  DOI: 10.1504/IJCAT.2024.10071229
   
  
• Location selection of intelligent warehouse combined with swarm intelligence  
  by Jinghui Wu, Junhua Yan 
  Abstract: The location selection of an intelligent warehouse is a critical decision that affects its efficiency and profitability. This paper introduces a new method that combines swarm intelligence with geographic information systems to tackle this challenge. The approach models the issue as a multi-objective optimization problem and employs a swarm intelligence algorithm to determine the best locations for setting up a warehouse. The algorithm considers several factors, including the proximity to customers, the availability of transportation infrastructure, and the cost of land and labor. To assess the efficacy of the proposed method, the algorithm was implemented in a real-world scenario involving a logistics company as a case study. The results demonstrate that the proposed approach can effectively identify optimal warehouse locations that improve the efficiency and profitability of the logistics operations.
  Keywords: location selection; intelligent warehouse; swarm intelligence; geographic information systems.
  DOI: 10.1504/IJCAT.2025.10071337
   
  
• MOOC system platform based on edge computing and artificial intelligence  
  by Bifeng Li, Lilibeth Cuison 
  Abstract: Online users flexibly obtain learning resources on MOOC system platform. Nowadays learners need to spend more time to screen the relevant content of the curriculum. Based on this, it aims to use artificial intelligence technology to assist online education. A large amount of video data brings high computing load and low real-time performance to the cloud server. Therefore, it is feasible to combine edge computing and artificial intelligence by taking advantage of the characteristics that the edge end has certain computing power and low latency near the terminal. This paper proposes a hybrid recommendation model Deep-AM based on artificial intelligence, and deploys it to the edge server. Thus, it can perform real-time detection and feature extraction. Compared with traditional models LFM, NARRE and DeepCoNN, it is shown that the proposed Deep-AM has faster response speed and higher task completion rate. And it has better practicability when applied to MOOC system.
  Keywords: edge computing; artificial intelligence; MOOC system; deep-AM model.
  DOI: 10.1504/IJCAT.2025.10071338
   
  
• A secure and reliable large scale online physical education solution based on deep learning for cloud computing  
  by Wenbo Song 
  Abstract: With the continuous expansion of online education scale, its existing problems have gradually emerged. Therefore, it is necessary to develop a secure and reliable large scale online physical education solution, cultivate higher-order thinking and guide students to learn effectively. This article studies cloud computing and deep learning models. And it proposes an improved model based on graph knowledge tracing(GKT). The improved model can better discover causal relationships and use network structure to track students' knowledge status. Finally, experiments are conducted on the improved model and the classical model, proving that the improved model based on GKT performs well. The improved model based on GKT outperforms other models and has superior convergence speed and accuracy. It has good practicality for large scale online physical education, helps to obtain interactive information of implicit knowledge in the scheme, and increases the richness of education.
  Keywords: online physical education; cloud computing; deep learning; GKT.
  DOI: 10.1504/IJCAT.2025.10071339
   
  
• Electrochemical energy storage power stations decision-making via digital twins and simulation-based data fusion  
  by Zhoubo Weng, Yimin Deng, Zhiyong Zhao, Shanshan Zhao 
  Abstract: The digital twin model for power station uses a dynamic three-dimensional representation to map the physical system and real-time data, encompassing monitoring control, state evolution, analysis, evaluation, and integrating real-time operational laws, power station analysis, and inference logic. This enables real-time monitoring, operational management, intelligent analysis, virtual inspection, and simulation training. Moreover, the joint Kalman filter is employed for data fusion to enhance the fusion effect of heterogeneous data from multiple sources within the digital twin-based electrochemical energy storage power station. Simulation results demonstrate promising performance in terms of fusion error and efficiency. By leveraging accurate data fusion, the proposed data-driven digital twin for electrochemical energy storage power stations offers several benefits, including improved accuracy, operational efficiency, proactive maintenance, real-time monitoring, enhanced system reliability, and safety. These advantages significantly contribute to optimizing the data fusion process in electrochemical energy storage power stations, ultimately leading to enhanced performance and decision-making.
  Keywords: electrochemical energy storage power stations; digital twins; data-driven decision; data fusion; joint Kalman filter.
  DOI: 10.1504/IJCAT.2025.10071340
   
  
• DRN-MCOA: image deblurring using deep residual network with modified coot optimisation algorithm  
  by Godekere Shivashankar Yogananda, Ananda Babu Jayachandra, Ahmed Alkhayyat, Dayananda Pruthviraja 
  Abstract: In this manuscript, a hybrid model is introduced for effective image deblurring. A Deep Residual Network (DRN) is implemented for reducing artificial traces that results in pleasant denoised images. Secondly, a Modified Coot Optimization Algorithm (MCOA) is incorporated with the DRN for selecting optimal kernel and threshold parameters. The exploitation and exploration ability of the MCOA is improved by employing an opposition based learning method and Cauchy mutation This process resolves the problem of local optima and improves the convergence rate. This DRN-MCOA model’s efficacy is investigated on real time images and RealBlur dataset. The DRN-MCOA model obtained a Peak Signal to Noise Ratio (PSNR) of 33.40 dB and a Structural Similarity Index (SSIM) of 0.96 on a real time collected image. Correspondingly, it achieved PSNR of 30.34 dB and 37.55 dB and SSIM of 0.92 and 0.96 on the RealBlur-J and RealBlur-R dataset.
  Keywords: coot optimisation algorithm; deep residual network; image deblurring; image processing; restoration.
  DOI: 10.1504/IJCAT.2025.10071789
   
  
• Research on image encryption algorithm based on Logistic chaotic system  
  by Qing Lu, Te Zhang, Junxiang Wan, Siyuan Xu 
  Abstract: For the recently proposed image encryption algorithm [1] based on the Logistic chaotic system, the algorithm first generates chaotic sequences and plaintext images through the key for XOR operations, and then pixel chaos and block chaos according to the chaotic sequence, and finally obtains the ciphertext image. Through analysis, it is found that the correlation between plaintext and ciphertext images in this algorithm is not high, and there is no ciphertext feedback mechanism. This algorithm can be attacked by selecting ciphertext. In this paper, the algorithm is improved. And the relevant test analysis of the improved algorithm is carried out, and the results are better than the algorithm proposed by Zhang and others.
  Keywords: chaotic system; image encryption; chosen-ciphertext attack.
  DOI: 10.1504/IJCAT.2025.10072127
   
  
• Hie-Graph-YOLOv9: a hierarchical YOLOv9 model with graph-based SE attention mechanism for vehicle detection in complex background
  by T. Selvamuthukumar, K. Vijayalakshmi, P. Dhanalakshmi, R. Abinaya 
  Abstract: Advanced vehicle detection algorithms are key to Intelligent Transportation Systems (ITS), enabling real-time traffic analysis, congestion and security management. Existing models like YOLOv9 face challenges in feature selection and learning, especially in dynamic or cluttered environments. To address these limitations, this research proposes Hie-Graph-YOLOv9 which is an extended version of YOLOv9 based on improving the feature selecting, feature learning and loss function by incorporating Hiera Transformers, Graph-based GAN-SE attention mechanism and Geometric-based Weighted Smooth L1 loss function. Hiera Transformers, integrated into the backbone network across four stages, refine multi-scale feature learning, ensuring robust representation of fine-grained and global patterns. The Graph-based GAN-SE, embedded in the bottleneck module, emphasises critical regions of feature maps, enhancing detection accuracy. Additionally, a Geometric-based Weighted Smooth L1 loss function is employed for bounding box regression, improving convergence speed and training stability. Experimental evaluations demonstrate the superiority of Hie-Graph-YOLOv9, achieving an AP (0.5) of 79.5%, improvement of faster convergence by 120 Epochs and an increased inference speed of 41.95 FPS, outperforming state-of-the-art models. This work offers a significant step forward in vehicle detection under complex real-world conditions.
  Keywords: object detection; YOLO; vehicle; Hiera; graph; squeeze and excitation.
  
  
• Hie-Graph-YOLOv9: A Hierarchical YOLOv9 model with Graph-based SE attention mechanism for vehicle detection in complex background  
  by T. Selvamuthukumar, K. Vijayalakshmi, P. Dhanalakshmi, R. Abinaya 
  Abstract: Advanced vehicle detection algorithms are key to intelligent transportation systems (ITS), enabling real-time traffic analysis, congestion and security management. The proposed Hie-Graph-YOLOv9 method is an extended version of YOLOv9 based on improving the feature selecting, feature learning and loss function. In this YOLO architecture, we induced Hiera Transformers in the backbone network in four stages for improving the feature learning. We also introduced the Graph based GAN-SE attention mechanism in the bottleneck module for giving attention to essential feature map regions and utilized Geometric based Weighted Smooth L1 loss function for bounding box prediction for faster convergence, training stability and improved accuracy.
  Keywords: object detection; YOLO; vehicle;Hiera; graph; Squeeze and Excitation.
  DOI: 10.1504/IJCAT.2025.10072853
   
  
• Evolutionary intelligent physical education mechanism in smart cities  
  by Liang Li, Fu Li, Nan Wang, Yang Liu 
  Abstract: The rapid evolution of smart cities has created unprecedented opportunities to revolutionize physical education through advanced technologies and data-driven optimization. This paper proposes an evolutionary intelligent physical education (EIPE) algorithm that employs surrogate-assisted evolutionary computation to optimize physical education programs in smart cities efficiently. EIPE incorporates diversity-based candidate generation, intelligent selection strategies, and adaptive surrogate model management to balance exploration and exploitation in the vast search space of program configurations. The algorithm's effectiveness is evaluated on benchmark problems specifically adapted to reflect real-world physical education scenarios. Experimental results demonstrate EIPE's superior performance to state-of-the-art approaches, achieving 27.8% better convergence and 35.9% improved solution diversity. The algorithm's ability to efficiently handle multiple competing objectives while maintaining solution diversity makes it particularly suitable for optimizing physical education programs in modern smart cities, where program adaptability and resource efficiency are crucial for promoting public health and well-being.
  Keywords: evolutionary computation; physical education; smart cities; surrogate-assisted optimisation; multi-objective optimisation.
  DOI: 10.1504/IJCAT.2025.10073194
   
  
• Metaverse-driven instrumental music teaching: a human-computer interactive perception method  
  by Tianyu Chen 
  Abstract: Teaching musical instruments traditionally requires one-on-one instruction, creating challenges of accessibility, cost, and feedback quality. Metaverse technologies offer promising solutions but face limitations in gesture recognition accuracy and haptic feedback realism. This paper proposes a metaverse-driven human-computer interaction method for instrumental music teaching that addresses these challenges. Through analysis of similar gestures in instrumental music performance, we discovered that inflection point count and gesture displacement vector angle significantly improve discrimination between similar instrumental gestures. By incorporating these features into our dynamic gesture recognition method, we enhance identification accuracy for specialized musical movements. Additionally, we simulate the haptic interaction process between virtual hands and instruments to achieve realistic performance effects and synchronous multi-channel sensory feedback in real-time interactions. Our approach focuses on six degrees of freedom haptic interactions during multi-point contact scenarios. Experimental results demonstrate that our method achieves higher recognition accuracy than baseline approaches for customized gestures and instrument-specific movements. This research contributes to more accessible, effective, and immersive instrumental music education in virtual environments, potentially democratizing access to quality music instruction regardless of geographical or economic constraints.
  Keywords: metaverse; instrumental teaching; human-computer interaction; VR; leap motion; six degrees of freedom.
  DOI: 10.1504/IJCAT.2025.10073195
   
  
• Large model driven adaptive English translation mechanism  
  by Haiying Song 
  Abstract: Significant advancements have been made in the field of natural language processing thanks to the development of large-scale language models. However, existing translation methods often face limitations in handling contextual nuances, domain-specific terminology, and evolving language usage. To address these challenges, we propose an adaptive English translation mechanism based on large models, incorporating adaptive feedback integration, domain-specific fine-tuning, and a dynamic learning mechanism. Our approach leverages transformer architectures to continuously refine translations by incorporating real-time user feedback and specializing in various domains. Evaluations conducted on datasets including WMT, OpenSubtitles, IWSLT, and CCMT demonstrate significant improvements in BLEU, METEOR, and TER scores compared to other models. Ablation studies confirm the contributions of each component, highlighting the importance of adaptive mechanisms in achieving superior translation performance. These findings indicate that our proposed method offers robust and adaptive solutions for diverse linguistic applications.
  Keywords: large model; natural language processing; dynamic learning; translation.
  DOI: 10.1504/IJCAT.2025.10073197
   
  
• Cross-cultural adaptation of English language and literature through context-augmented neural machine translation  
  by Zhihua Duan 
  Abstract: The cross-cultural adaptation of the English language and literature plays a crucial role in fostering global communication and understanding. This paper proposes a novel approach to enhance document-level neural machine translation (NMT) models by incorporating a hierarchical global context derived from the entire document. Specifically, the proposed model obtains the dependencies between the word in the current sentence and all sentences and words in the document, respectively, and combines the dependencies at different levels to obtain the global context representation containing hierarchical contextual information. Finally, each word in the current sentence of the source language acquires its unique context that integrates word-and sentence-level dependencies. This paper proposes a two-step training strategy to use the advantages of parallel sentence pairs in training fully. Experiments on several benchmark corpus datasets show that the proposed model achieves significant translation quality improvement compared with several strong baselines.
  Keywords: cross-cultural adaptation; English language and literature; context augmented; neural machine translation.
  DOI: 10.1504/IJCAT.2025.10073212
   
  
• Neural networks driven differentiation analysis of art painting works  
  by Lei Xia 
  Abstract: Fine art paintings are a significant part of cultural heritage, reflecting human creativity and emotional expression through visual representation. The analysis of paintings using artificial intelligence techniques plays a crucial role in both the preservation and innovative development of artistic works. Brushstrokes, as fundamental elements of paintings, carry stylistic and expressive characteristics that are essential for understanding artistic styles. This paper designs an image classification method for paintings that integrates global and local features, where a convolutional neural network is used to obtain an overall stylistic description of an image of a painting. To verify the effectiveness, this paper retrieves painting data from painting databases and the web and constructs a database containing 2040 painting images. The experiments show that the method proposed can effectively analyze the differences of fine art paintings and achieve classification and distinction.
  Keywords: art painting classification; brushstrokes; CNN; LSTM; attention mechanism.
  DOI: 10.1504/IJCAT.2025.10073213
   
  
• Eco-friendly interior design mechanism with AI and big data analysis  
  by Jing Wang 
  Abstract: The interior design practices can promote overall sustainability in urban environments by leveraging artificial intelligence (AI) and advanced data analytics techniques. In this paper, a smart eco-friendly interior design optimization approach is proposed based on wearable sensor data analysis. Meanwhile, real-time analysis of user behavior is performed using sensor data from wearable devices. The gated recurrent unit identifies user behavior, establishes a correlation between user behavior and comfort requirements, updates user comfort preferences, and dynamically estimates model parameters based on environmental sensor data. Subsequently, a model predictive control-based solution method for optimizing energy efficiency in smart interior design is introduced. Experiments are conducted involving four typical user behavior scenarios, demonstrating improvements in the economic and comfort aspects of smart, eco-friendly interior design. Adopting model predictive control-based optimization decisions in eco-friendly interior design solutions within smart cities offers numerous benefits.
  Keywords: eco-friendly interior design; artificial intelligence; big data; model predictive control.
  DOI: 10.1504/IJCAT.2025.10073214
   
  
• A layout system design for ship engine room and equipment by artificial intelligence  
  by Meijing Song, Karia Noorliza 
  Abstract: This study aims at the problems of traditional intelligent algorithms, such as being prone to getting trapped in local optima, having slow convergence speeds, and insufficient accuracy when solving complex optimisation problems. It systematically compares and analyses the advantages and limitations of the genetic algorithm (GA), particle swarm optimisation (PSO) algorithm, and ant colony algorithm (ACA). Moreover, a GA-PSO-ACA that combines the advantages of the three is proposed. Taking the layout design of ship engine room equipment as the research object, under the consideration of multiple constraints, such as no overlap of equipment, balanced weight distribution, operation space, and safety distance, three traditional algorithms and the GA-PSO-ACA are respectively used for simulation optimisation. The results show that the optimal objective function value of the GPA algorithm is 160, which is significantly better than that of the GA (220), the PSO algorithm (225), and the ACA (218).
  Keywords: ship design; genetic algorithm; particle swarm optimization; ant colony algorithm; hybrid algorithm.
  DOI: 10.1504/IJCAT.2025.10073304
   
  
• Graphic design optimisation mechanism based on deep learning in smart cities  
  by Yinan Chen 
  Abstract: This article focuses on the background of smart cities, analyzes and optimizes urban graphic design based on deep learning, and proposes the improved UNet model based on the coordinate attention(CA-IUN). First, we improve the model based on UNet. The improved UNet model (IUN) replaces some traditional convolutions in the encoding and decoding stages with dilated convolutions. Then, transposed convolution is used for upsampling, replacing traditional linear interpolation. We also design multi scale fusion using phantom convolution and SENet. CA-IUN adds coordinate attention module to the encoder and decoder of IUN to focus on the specific positions of features. In addition, this article combines perceptual loss and smooth L1 loss function to train the network. Finally, experiments are shown that CA-IUN outperforms other models in optimizing graphic design, indicating that CA-IUN can effectively achieve more refined and efficient graphic design optimization in smart cities.
  Keywords: deep learning; graphic design optimisation; smart cities.
  DOI: 10.1504/IJCAT.2025.10073305
   
  
• Adaptive constraint multi-objective evolutionary computation industrial economic optimisation in smart city  
  by Yao Lv, Zimeng Guo 
  Abstract: This paper introduces an adaptive constraint multi-objective evolutionary algorithm for smart city industrial economics (ACMEA-SCIE). ACMEA-SCIE employs a dual reproduction strategy, evolving two complementary populations: a main population for exploring diverse industrial configurations and an archive population for preserving high-quality solutions. Additionally, a dynamic fitness allocation function adaptively balances objective optimization and constraint handling, while an innovative archive update mechanism maintains solution diversity. The algorithm's performance was evaluated on three benchmark sets: smart city resource allocation, industrial ecosystem optimization, and dynamic urban industrial planning. Experimental results demonstrate ACMEA-SCIE's superior performance compared to state-of-the-art algorithms, achieving significant improvements in both inverted generational distance and hypervolume metrics. Additional analyses, including convergence performance and solution distribution, further validate ACMEA-SCIE's effectiveness. The proposed algorithm shows remarkable adaptability across various problem types, enhanced constraint handling, and improved multi-objective balancing.
  Keywords: smart city; industrial economic; evolutionary computation; multi-objective optimisation.
  DOI: 10.1504/IJCAT.2025.10073306
   
  
• Sports injury prediction based on sensor information fusion and neural network  
  by Ying Song 
  Abstract: A sensor information fusion method for sports injury prediction is proposed in this paper. The hole effect is eliminated by employing the accumulation of multi-frame differences. On this basis, accurate motion regions are determined by fusion sensors to monitor motion in different scenes. Non-stationary signals of monitoring results are analyzed by wavelet analysis method to obtain motion injury characteristics. Machine learning algorithms can be trained on this sensor data to develop predictive models for sports injuries. Sensor information fusion and wavelet radial basis function neural network are combined to obtain the wavelet eigenvector of all sensors. A radial basis function neural network will output a value when the data sent to it matches a certain risk level to achieve sports injury prediction. The results reveal that the proposed model performs well in prediction accuracy and running time, which can provide real-time feedback to athletes and coaches.
  Keywords: sports injury; sensor information fusion; RBF; wavelet; Neural network.
  DOI: 10.1504/IJCAT.2025.10073307
   
  
• Evaluation of ultra-large-scale English translation mechanism based on Bi-LSTM  
  by Yafei Bi 
  Abstract: How to effectively extract and utilize syntactic features in the model is an issue worthy of further study in the current translation quality estimation task. This paper introduces a Bi-directional Long Short-Term Memory (Bi-LSTM) based English translation mechanism evaluation model aimed at providing fast and accurate feedback to enhance machine translation systems. The proposed model incorporates the following strategies. Firstly, we utilize the Skip-gram model and the Continuous Bag of Words (CBOW) model of the Word2vec to preprocess text data before feature extraction. Second, we utilize three types of translation feature to promote the performance of translation evaluation, including word prediction feature, word-embedding feature, and syntactic structure feature. Third, we design a English translation mechanism evaluation model based on the Bi-LSTM model by fusing the three types of extracted features. The results of the experiment demonstrate that the approach suggested in this paper exhibits favorable evaluation performance.
  Keywords: machine learning; English translation; evaluation model; neural network; feature extraction.
  DOI: 10.1504/IJCAT.2025.10073308
   
  
• Design of library information recommendation system integrating transfer learning and population intelligence optimisation  
  by Fang He 
  Abstract: Traditional classification and recommendation methods encounter limitations due to high dimensionality, insufficient annotation, and the heterogeneous nature of user interests. To address these challenges, this study proposes a novel framework for user interest classification and book recommendation, named Transfer Convolutional Adaptive Support Vector Machine (T-CASVM). This framework integrates deep transfer learning with particle swarm optimization (PSO). It utilizes a deep transfer convolutional neural network with shared weights to extract features from both source and target domains, thereby mitigating distributional discrepancies by simultaneously optimizing classification loss and domain loss. Furthermore, PSO is employed to refine the classifier, improving both accuracy and computational efficiency. The framework calculates cosine similarity between the target user and others to provide personalized book recommendations. Experimental results on the public Book-Crossing dataset show that T-CASVM outperforms traditional methods, achieving over 0.79 in precision index concerning the user interest classification task.
  Keywords: swarm intelligence; PSO; transfer learning; recommendation system.
  DOI: 10.1504/IJCAT.2025.10073345
   
  
• A large-scale high-definition music performance strategy based on the combination of reality and metaverse  
  by Minglong Wang, Manqi Kongshi, Daohua Pan 
  Abstract: In this paper, we explore the intersection of the Metaverse, music generation, deep learning, and performance strategy. Deep learning techniques have shown promise in generating music, and can be applied to create personalized soundscapes for users in the Metaverse. However, creating music with deep learning is a complex process that requires careful consideration of performance strategy. Factors such as data quality, model selection, and training methodology can significantly influence the quality of generated music.In this paper, we propose a method for large-scale high-definition music generation and dance performance by combining Metaverse and deep learning techniques. First, we use the transformer model to generate polyphonic music. Then, we use the VAE to encode dance movements. Finally, we use a joint attention mechanism to map music to dance performances. Experimental results and comparative analysis show the effectiveness of the proposed method.
  Keywords: reality and metaverse; deep learning; music generation; large-scale creation.
  DOI: 10.1504/IJCAT.2025.10073494
   
  
• Preschool education video image optimisation mechanism based on deep evolutionary learning in smart city  
  by Junqing Fan 
  Abstract: As an important stage of basic education, the richness and quality of teaching resources in preschool education directly affect the growth and development of children. In order to better optimize the processing of preschool education video image and improve their clarity, this paper proposes a deep evolutionary learning method based on the improved whale optimization algorithm and bi-directional long-short term memory (IWOA-BiLSTM). BiLSTM utilizes the temporal information between adjacent frames of preschool education video image to preserve the time series output in the feature map of preschool education video image. This can fully learn the information between adjacent frames of preschool education video image, making the optimized image contain richer information. IWOA is used to optimize the key parameters of BiLSTM and improve its optimization performance. Finally, experiments show that IWOA-BiLSTM can effectively optimize preschool education video image in smart city.
  Keywords: deep learning; image optimisation; evolutionary algorithms; preschool education video; smart city.
  DOI: 10.1504/IJCAT.2025.10073495
   
  
• Digital media art design mechanism based on reinforcement learning in smart city  
  by Xin He 
  Abstract: Digital media art has emerged as a pivotal domain that intersects technology, culture, and urban life, transforming public spaces and offering novel forms of interaction and expression. In this paper, we presents a novel framework for 3D face reconstruction in digital media art design, leveraging Generative Adversarial Networks (GANs) and Reinforcement Learning (RL). We train and evaluate our model with rigorous experiments based on public dataset, comparing its performance against several state-of-the-art methods. Our proposed model demonstrates superior performance in two metrics. Additionally, we conduct convergence analysis and robustness to input noise experiments to further validate our approach. The results highlight the effectiveness of our method in producing high-quality, realistic, and robust 3D face reconstructions, underscoring its potential for enhancing digital media art installations in smart cities.
  Keywords: digital media; smart city; reinforcement learning.
  DOI: 10.1504/IJCAT.2025.10073496