检索结果-内蒙古大学图书馆

Synthesis of batik motifs using a diffusion - generative adversarial network

Multimedia Tools and Applications 2025年第7期84卷 3407-3438页

作者： Octadion, One Yudistira, Novanto Kurnianingtyas, Diva Informatics Engineering Faculty of Computer Science Universitas Brawijaya Malang Indonesia

This paper presents a novel approach for generating intricate Batik motifs using a modified Diffusion-Generative Adversarial Network (Diffusion-GAN) augmented with StyleGAN2-Ada. Motivated by the rich cultural heritage of Indonesian Batik, our research addresses the challenge of synthesizing high-quality, diverse patterns that capture the artistry and complexity of traditional designs. Traditional generative models often struggle with stability and fidelity in artistic synthesis. We integrate StyleGAN2-Ada and Diffusion techniques to overcome these limitations, optimizing model architecture and employing a curated Batik dataset. Evaluation metrics including Frechet Inception Distance (FID), Kernel Inception Distance (KID), precision, recall, and non-redundancy assess the quality and diversity of generated motifs. Our results demonstrate significant advancements in the realism and authenticity of synthesized Batik patterns, leveraging Diffusion-GAN to enhance detail and artistic fidelity. By exploring the intersection of deep learning and cultural art, this research contributes to the preservation and evolution of Batik as a cultural artifact. References to prior works underscore the evolution of Artificial Intelligence (AI) in cultural heritage preservation, highlighting the potential of modern AI techniques to innovate within traditional art forms like Batik. In summary, this study showcases the efficacy of the proposed Diffusion-GAN methodology in synthesizing original and convincing Batik motifs, while identifying opportunities for further refinement and complexity in motif synthesis. The dataset and code can be accessed here: https://***/octadion/diffusion-stylegan2-ada-pytorch © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing demand forecasting through combination of anomaly detection and continuous improvement

引用

Soft Computing 2025年第2期29卷 1243-1258页

作者： Jahani, Meysam Zojaji, Zahra Raji, Fatemeh Faculty of Computer Engineering University of Isfahan Isfahan Iran School of Computer Science and Informatics De Montfort University Leicester United Kingdom

Demand forecasting has emerged as a crucial element in supply chain management. It is essential to identify anomalous data and continuously improve the forecasting model with new data. However, existing literature fails to comprehensively cover both aspects of anomaly detection and continuous improvement in demand forecasting. This study proposes an enhanced model to improve accuracy in the demand forecasting. The proposed model introduces a novel data handling method that incorporates an anomaly detection autoencoder, improved with anomaly correction mechanisms. The data handling approach simultaneously detects data anomalies, distinguishes between expected and unexpected anomalies, and corrects anomalous data, ensuring cleaner input for demand forecasting. Then, the proposed model employs a long short-term memory architecture for demand forecasting, enhanced with a continuous improvement method. Thus, the model not only forecasts demand but also retrains the model when the anomaly data surpasses the predetermined threshold, thereby improving the accuracy of forecasting. The results show that the proposed model outperforms other models in detecting data anomalies, achieving an average precision-recall of 0.922, a receiver operating characteristic value of 0.739, and a significance level of less than 0.05. Finally, the model exhibits superior performance in demand forecasting, with average mean squared error, root mean squared error, and mean absolute error values of 33.167, 4.347, and 1.509, respectively, all with a significance level of less than 0.05. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Demand response

来源：评论

学校读者我要写书评

暂无评论

An effective intrusion detection system based on feature selection and regularized long short-term memory classifier

引用

Multimedia Tools and Applications 2025年 1-24页

作者： Chikkalwar, Sudha Rani Garapati, Yugandhar Department of Computer science and Engineering GITAM Deemed to be University Rudraram Hyderabad India Department of Computer science and Informatics Mahatma Gandhi University Telangana Nalgonda India

In recent days, the expansion of Internet of Things (IoT) and the quick advancement of computer system applications contribute to the current phenomenon of data growth. The field of intrusion detection has expanded considerably as a result of the need to protect networks from various attacks. Previously, security has been provided by combining conventional intrusion monitoring systems with a high number of security measures. However, the ability of combined systems to analyze enormous amounts of data has been restricted. Therefore, in this research, open-source and freely available datasets known as NSL-KDD and CSE-CIC-IDS2018 datasets are analyzed. Those datasets contain a class column that splits the attack detection into two classes (normal and anomalous) and four categories (DoS, probe, R2L, and U2R). Then, feature encoding and scaling are used for pre-processing. After that, a SelectPercentile-based feature selection algorithm is suggested to identify the numerous various intrusion attacks and choose the best features. Later, the Regularized Long Short-Term Memory (LSTM) is employed to classify the data. The result analysis verified that Regularized LSTM surpasses the existing models in the NSL-KDD based on accuracy (99.95%), false alarm rate (0.05%), attack detection rate (99.82%), and f1-measure (99.94%). While in the CSE-CIC-IDS2018 dataset, the Regularized LSTM method achieves better accuracy (99.33%), false alarm rate (0.07%), attack detection rate (99.96), and f1-measure (99.95%). © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

3D-SCUMamba: An Abdominal Tumor Segmentation model

引用

IEEE Access 2025年 13卷 98362-98373页

作者： Juwita Hassan, Ghulam Mubashar Datta, Amitava Departement of Computer Science and Software Engineering The University of Western Australia Australia Departemet of Informatics University of Syiah Kuala Indonesia

Identification and segmentation of tumors from CT scans are essential for early detection and effective treatment but they remain challenging due to imaging artifacts and significant variability in tumor location, size, and morphology. Existing deep learning models typically adopt encoder-decoder architectures integrating convolutional layers with global dependency modeling to capture broader contextual information around tumors. Capturing global dependencies is critical, as local information alone often inadequately distinguishes tumors, especially given their diverse sizes, complex shapes, and interference from imaging artifacts. Recent advancements observe the emergence of the convolution-transformer architecture, which improves segmentation performance on the cost of substantial computational resources. In this article, we propose 3D-SCUMamba, an innovative architecture strategically integrating State Space Modeling- Based deep learning (Mamba) within the bottleneck of the encoder-decoder structure to overcome the limitations of existing segmentation networks. The proposed model efficiently models global dependencies while maintaining stable training dynamics and efficient inference. Additionally, we introduce a novel Spatio-Context (SC) module utilizing 3D convolutions without pooling to enhance feature representations and reduce information loss commonly associated with pooling operations. The SC module effectively prepares summarized features for robust global dependency processing by the Mamba component. 3D-SCUMamba explicitly addresses the prevalent limitations in current deep learning methods by prioritizing robust feature representation, training stability, and substantial accuracy *** evaluations conducted on three medical imaging datasets: MSD Pancreas Tumor, MSD Colon Tumor, and Synapse BTCV, demonstrate that computationally efficient 3D-SCUMamba consistently outperforms state-of-the-art methods with segmentation accuracy improve

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

DiverseNet: Decision Diversified Semi-supervised Semantic Segmentation Networks for Remote Sensing Imagery

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 14258-14271页

作者： Ma, Wanli Karakus, Oktay Rosin, Paul L. Cardiff University School of Computer Science and Informatics Cardiff United Kingdom University of Cambridge Department of Engineering Cambridge United Kingdom

Semi-supervised learning (SSL) aims to help reduce the cost of the manual labelling process by leveraging a substantial pool of unlabelled data alongside a limited set of labelled data during the training phase. Since pixel-level manual labelling in large-scale remote sensing imagery is expensive and time consuming, semi-supervised learning has become a widely used solution to deal with this. However, the majority of existing SSL frameworks, especially various teacher-student frameworks, are too bulky to run efficiently on a GPU with limited memory. There is still a lack of lightweight SSL frameworks and efficient perturbation methods to promote the diversity of training samples and enhance the precision of pseudo labels during training. In order to fill this gap, we proposed a simple, lightweight, and efficient SSL architecture named DiverseHead, which promotes the utilisation of multiple decision heads instead of multiple whole networks. Another limitation of most existing SSL frameworks is the insufficient diversity of pseudo labels, as they rely on the same network architecture and fail to explore different structures for generating pseudo labels. To solve this issue, we propose DiverseModel to explore and analyse different networks in parallel for SSL to increase the diversity of pseudo labels. The two proposed methods, namely DiverseHead and DiverseModel, both achieve competitive semantic segmentation performance in four widely used remote sensing imagery datasets compared to state-of-the-art semi-supervised learning methods. Meanwhile, the proposed lightweight DiverseHead architecture can be easily applied to various state-of-the-art SSL methods while further improving their performance. The code is available at Here. © 2008-2012 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Predicting Student Academic Performance Using an Optimised Regression Machine Learning Model 9

Predicting Student Academic Performance Using an Optimised R...

引用

9th IEEE International Conference on Adaptive science and Technology, ICAST 2024

作者： Martey, Ezekiel Mensah Abdullah, Abubakar Appiah, Obed Department of Computer Science and Engineering Tarkwa Ghana Department of Computer Science and Informatics Sunyani Ghana

ISBN: (纸本)9798350385403

Machine learning has emerged as one transformative tool in predicting student academic performance. This study evaluates machine learning models regression in predicting student academic performance. It can be found optimised Support Vector Regression (SVR) model accurately predicts student academic performance using a benchmark dataset from Kaggle. The approach incorporated Recursive Feature Elimination (RFE) for optimal feature selection while key parameters of the SVR model are tuned through grid search optimisation to enhance model performance. The optimised SVR model achieves exceptional predictive accuracy with an R-squared value of 0.97, Mean Absolute Error of 2.98, Mean Squared Error of 13.28, Root Mean Squared Error of 3.64, Median Absolute Error of 2.65, and Mean Absolute Percentage Error of 0.23. Comparative analysis demonstrates the superior performance of optimised SVR model against Decision Tree, Random Forest, K-Nearest Neighbors, Bayesian Ridge, and Gradient Boosting models. Experimental results demonstrate the validity of robustness and reliability of our approach, offering valuable insights for educational institutions to develop targeted interventions and support systems. © 2024 IEEE.

关键词： Educational Data Mining Grid Search Optimization Machine Learning Recursive Feature Elimination Student Performance Prediction

来源：评论

学校读者我要写书评

暂无评论

CBIR framework for Diagnosing COVID 19 in Chest X-Ray Images using Deep learning by integrating Photometric Data Augmentation 9

CBIR framework for Diagnosing COVID 19 in Chest X-Ray Images...

引用

9th IEEE International Conference on Adaptive science and Technology, ICAST 2024

作者： Martey, Ezekiel Mensah Appiah, Obed Sitti, Mensah Nusetor, Foster Delali Department of Computer Science and Engineering Tarkwa Ghana Department of Computer Science and Informatics Sunyani Ghana

ISBN: (纸本)9798350385403

computer-Aided Diagnostic (CAD) systems enhance clinical decision-making, providing more accurate and streamlined processes. CAD systems based on Content-Based Image Retrieval (CBIR) serve as visual tools that strengthen the reliability of clinical decisions. While deep learning-based CAD systems have shown significant progress in extracting insights from X-rays and medical images, they often struggle due to their reliance on large datasets and high computational demands. These challenges are particularly acute in the medical domain, where datasets are frequently limited and imbalanced. This paper presents a framework for diagnosing COVID-19 from X-ray images, combining deep learning with photometric data augmentation. The proposed approach focuses on two core tasks: image retrieval and classification, which depend on robust feature representation. Using Euclidean distance measures, we utilise a modified EfficientNet-B3 model for feature extraction and examine the effects of transfer learning and fine-tuning on retrieval and diagnostic accuracy. Evaluating binary and multiclass datasets reveals that the proposed method performs strongly in classification tasks, with notable effectiveness in binary and multiclass scenarios. Compared to state-of-the-art techniques, our approach demonstrates superior accuracy for binary classification of X-ray images, outperforming most existing models. © 2024 IEEE.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Concentric Blocking Techniques for Improved Feature Extraction in Local Binary Pattern (LBP) Systems 9

Concentric Blocking Techniques for Improved Feature Extracti...

引用

9th IEEE International Conference on Adaptive science and Technology, ICAST 2024

作者： Appiah, Obed Martey, Ezekiel Mensah Department of Computer Science and Informatics Sunyani Ghana Department of Computer Science and Engineering Tarkwa Ghana

ISBN: (纸本)9798350385403

The domain of image analysis has seen rapid advancements, particularly feature extraction techniques, driven by the increasing demand for accurate and efficient results. Among these advancements, Local Binary Patterns (LBP) have gained prominence due to their simplicity and computational efficiency. However, traditional LBP methods often need help with image orientation and scale variations, common in real-world scenarios. To address these challenges, this study explores the potential of concentric blocking methods as an alternative to conventional grid-based blocking techniques. Concentric blocking, which organizes image data into circular or spiral patterns, is hypothesized to enhance the robustness and accuracy of LBP-based systems by better accommodating geometric transformations. The study uses multiple LBP variants to compare traditional grid-based blocking with concentric blocking techniques across various image analysis tasks, including face recognition and texture classification. Experimental results demonstrate that the proposed ring-centric technique significantly outperforms traditional methods in precision and computational efficiency. Precision values for the proposed Ring Centric technique reach up to 90.73% for the basic LBP and 88.47% for Neighboring LBP with Distance (NLBPD), indicating its superior performance in handling variations in orientation and scale. Additionally, the technique demonstrates lower running times, making it a highly efficient and accurate choice for image retrieval tasks. This study highlights the importance of image-blocking strategies in enhancing the performance of LBP-based systems and underscores the potential of concentric blocking methods in overcoming the limitations of traditional approaches. The findings contribute to developing more resilient and efficient image analysis systems, with significant implications for applications in computer vision, medical diagnostics, and security systems. © 2024 IEEE.

关键词： Local binary pattern

来源：评论

学校读者我要写书评

暂无评论

A NEW STEPSIZE STRATEGY FOR KORPELEVICH’S ALGORITHM SOLVING VARIATIONAL INEQUALITIES

Applied Set-Valued Analysis and Optimization

引用

Applied Set-Valued Analysis and Optimization 2024年第3期6卷 309-317页

作者： Trinh, Ngoc Hai Tran, Trung Duc Faculty of Mathematics and Informatics Hanoi University of Science and Technology Hanoi Viet Nam College of engineering and computer science VinUniversity Hanoi Viet Nam

In this paper, we propose two self-adaptive extragradient-like algorithms for solving pseudomonotone variational inequalities. We consider two cases: the mapping is Lipschitz continuous (with unknown modulus) and is not Lipschitz continuous. The step size in our algorithms can either increase or decrease at each step. This feature makes our algorithms more efficient. We also give some numerical examples to compare the performance of our algorithm with the existing one. ©2024 Applied Set-Valued Analysis and Optimization.

关键词： Variational techniques

来源：评论

学校读者我要写书评

暂无评论

Integration of data science with the intelligent IoT(IIoT):Current challenges and future perspectives

引用

Digital Communications and Networks 2025年第2期11卷 280-298页

作者： Inam Ullah Deepak Adhikari Xin Su Francesco Palmieri Celimuge Wu Chang Choi Department of Computer Engineering Gachon UniversitySeongnamSujeong-gu 13120Republic of Korea University of Electronic Science and Technology of China Chengdu 611731China College of Information Science and Engineering Hohai University213000China Department of Computer Science University of SalernoItaly Graduate School of Informatics and Engineering The University of Electro-Communications 1-5-1ChofugaokaChofu-shiTokyo182-8585Japan

The Intelligent Internet of Things(IIoT)involves real-world things that communicate or interact with each other through networking technologies by collecting data from these“things”and using intelligent approaches,such as Artificial Intelligence(AI)and machine learning,to make accurate *** science is the science of dealing with data and its relationships through intelligent *** state-of-the-art research focuses independently on either data science or IIoT,rather than exploring their ***,to address the gap,this article provides a comprehensive survey on the advances and integration of data science with the Intelligent IoT(IIoT)system by classifying the existing IoT-based data science techniques and presenting a summary of various *** paper analyzes the data science or big data security and privacy features,including network architecture,data protection,and continuous monitoring of data,which face challenges in various IoT-based *** insights into IoT data security,privacy,and challenges are visualized in the context of data science for *** addition,this study reveals the current opportunities to enhance data science and IoT market *** current gap and challenges faced in the integration of data science and IoT are comprehensively presented,followed by the future outlook and possible solutions.

关键词： Data science Internet of things(IoT) Big data Communication systems Networks Security Data science analytics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：