检索结果-内蒙古大学图书馆

Deep hierarchical network for full-reference omnidirectional image quality assessment

Multimedia Systems 2025年第3期31卷 1-15页

作者： Zhang, Youzhi Wan, Lifei Chen, Yiren Zhou, Xiaofei Zhang, Xiaolin Liu, Deyang School of Computer and Informatio Anqing Normal Universit Anqing246000 China College of Electrical Engineering and Automation Shandong University of Science and Technology Qingdao266590 China School of Automation Hangzhou Dianzi University Hangzhou310061 China

Virtual Reality(VR) can provide users immersive experience, which is considered as an innovative teaching tool for education. The rapid advancement of VR technology has brought a substantial increase in the availability of omnidirectional images on the internet, which highlights the urgent need for omnidirectional image quality assessment. Assessing the quality of omnidirectional images poses a significant challenge due to their spherical representation. Unlike traditional 2D images, omnidirectional images record 360∘×180∘ panoramic content with extremely high resolution, which brings more difficulties in handling the complicated scene content during quality assessment procedure. To relieve this problem, in this paper, we propose to learn a deep hierarchical network for full-reference omnidirectional image quality assessment. Motivated by the characteristic of the human visual system (HVS), the proposed method utilizes intermediate layers of a convolutional neural network (CNN) to extract multi-level information. These layers capture different semantic representations, encompassing both low-level and high-level details. The shallow layers capture low-level information, such as edges and corners, which are highly sensitive to quality degradation. To effectively model the impact of distortions, we further introduce a local distortion-aware (LDA) block which can accurately capture the degradation of image quality. The proposed method can achieve promising performances in experiments conducted on two publicly available databases, which demonstrates the effectiveness of our method. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Damage detection

来源：评论

学校读者我要写书评

暂无评论

From Misinformation to Truth: Fake News Detection with Transformer-Based Models

From Misinformation to Truth: Fake News Detection with Trans...

引用

International Conference on Communication Systems and Network Technologies (CSNT)

作者： Arvind Mewada Mohd. Aquib Ansari Sushil Kumar Maurya Computer Science & Engineering Bennett University Greater Noida India Computer Science & Engineering Siksha ‘O’ Anusandhan Bhubaneswar India

ISBN: (数字)9798331531935

ISBN: (纸本)9798331531942

Nowadays, technological advancements have significantly transformed information dissemination and consumption strategies. People acquire news through online portals such as social media and news websites. However, this change is also accompanied by the rapid spread of fake news based on false and misleading information. The main reason is that the coverage of digital platforms and social media is wide, and news can spread quickly without any authenticity verification. This problem poses a serious challenge to social trust, the ability to make wise decisions, and democratic stability. Therefore, in this study, we developed an advanced artificial intelligence (AI) model to detect fake news using Transformer (BERT, DistilBERT, and RoBERTa ) models and evaluated their performance based on indicators such as accuracy, recall, and F1 score. The results show that RoBERTa performs best due to the use of a large dataset and better contextual understanding and achieved an accuracy of 93.84%, a precision of 91.80%, a recall of 95.28%, and an F1 score of 90.82%.

关键词： Support vector machines Accuracy Social networking (online) Communication systems Machine learning Transformers Stability analysis Fake news Portals

来源：评论

学校读者我要写书评

暂无评论

Deep Networks as Denoising Algorithms: Sample-Efficient Learning of Diffusion Models in High-Dimensional Graphical Models

引用

IEEE Transactions on Information Theory 2025年第4期71卷 2930-2954页

作者： Mei, Song Wu, Yuchen University of California at Berkeley Department of Statistics Department of Electrical Engineering and Computer Sciences BerkeleyCA94720 United States University of Pennsylvania Wharton School Department of Statistics and Data Science PhiladelphiaPA19104 United States

We investigate the efficiency of deep neural networks for approximating scoring functions in diffusion-based generative modeling. While existing approximation theories leverage the smoothness of score functions, they suffer from the curse of dimensionality for intrinsically high-dimensional data. This limitation is pronounced in graphical models such as Markov random fields, where the approximation efficiency of score functions remains unestablished. To address this, we note score functions can often be well-approximated in graphical models through variational inference denoising algorithms. Furthermore, these algorithms can be efficiently represented by neural networks. We demonstrate this through examples, including Ising models, conditional Ising models, restricted Boltzmann machines, and sparse encoding models. Combined with off-the-shelf discretization error bounds for diffusion-based sampling, we provide an efficient sample complexity bound for diffusion-based generative modeling when the score function is learned by deep neural networks. © 1963-2012 IEEE.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

Exploration of soliton structures and modulation instability analysis for the highly dispersive perturbed NLSE with sextic-power law refractive index

引用

Optical and Quantum Electronics 2025年第1期57卷 1-18页

作者： Abdullah, Eman H. M. Ahmed, Hamdy M. Zaghrout, Afaf A. S. Bahnasy, Amal Ibrahim Ahmed Rabie, Wafaa B. Hashemi, M.S. Bayram, M. Department of Physics and Engineering Mathematics Higher Institute of Engineering El Shorouk Academy Cairo Egypt Department of Mathematics Faculty of Science Al Azhar University Cairo Egypt Department of Physics and Engineering Mathematics Higher Institute of Engineering and Technology Tanta Egypt Department of Mathematics Basic Science Faculty University of Bonab Bonab Iran Computer Engineering Biruni University Istanbul Turkey

This study introduces some novel soliton solutions and other analytic wave solutions for the highly dispersive perturbed nonlinear Schrödinger equation with generalized nonlocal laws and sextic-power law refractive index. This equation models the behavior of complex wave propagation. The improved modified extended tanh function governs the transmission of various solitons with different effects. New optical solitons are constructed as bright solitons, dark solitons, combo bright-dark solitons and singular solitons. Furthermore, exponential solutions, triangular solutions and rational wave solutions are shown. The findings conferred in this article show a crucial advancement in understanding the propagation of waves in non-linear media. The study also includes modulation instability analysis to examine the stability of the derived wave solutions, supplemented by graphical simulations that depict their distinctive features. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Solitons

来源：评论

学校读者我要写书评

暂无评论

Intelligent network intrusion detection system using optimized deep CNN-LSTM with UNSW-NB15

引用

International Journal of Information Technology (Singapore) 2025年 1-17页

作者： Thaljaoui, Adel Department of Computer Science and Information College of Science at Zulfi Majmaah University Al-Majmaah 11952 Saudi Arabia Preparatory Institute for Engineering Studies of Gafsa Street Al Khayzourane-Zarroug Gafsa 2112 Tunisia

In the era of interconnected technologies, the ongoing threat of intrusion attacks poses significant risks, exposing individuals and organizations to potentially severe repercussions. Network Intrusion Detection System (IDS) has a crucial role in securing these networks. In this paper, we propose an hybrid model combining the convolution neural network (CNN) and the long short-term memory (LSTM) with hyper-parameters optimization using Bayesian optimization method for attacks detection in IoT. We train and test our hybrid model using the benchmark dataset UNSW-NB15 for evaluating its effectiveness in intrusion detection. The performance of our model is assessed using various metrics, including accuracy, precision, recall, and F1-score. In addition, we conduct a comparative study to highlight the effectiveness of our proposed model, which achieved remarkable accuracy on the testing set. This evaluation underscores the model's potential for attacks detection. © Bharati Vidyapeeth's Institute of computer Applications and Management 2025.

关键词： Convolutional neural network (CNN) Deep learning Internet of Things (IoT) Intrusion detection system (IDS) Long short-term memory (LSTM) Optimization

来源：评论

学校读者我要写书评

暂无评论

Online Influence Maximization With Semi-Bandit Feedback Under Corruptions

引用

IEEE Transactions on Network science and engineering 2025年第3期12卷 2308-2321页

作者： Cheng, Xiaotong Nourani-Koliji, Behzad Maghsudi, Setareh Ruhr-University Bochum Department of Electrical Engineering and Information Technology Bochum44801 Germany University of Tübingen Department of Computer Science Tübingen72074 Germany Fraunhofer Heinrich Hertz Institute Berlin10587 Germany

In this article, we investigate the online influence maximization in social networks. Most prior research studies on online influence maximization assume that the nodes are fully cooperative and act according to their stochastically generated influence probabilities on others. In contrast, we study the online influence maximization problem in the presence of some corrupted nodes whose damaging effects diffuse throughout the network. We propose a novel bandit algorithm, CW-IMLinUCB, which robustly learns and finds the optimal seed set in the presence of corrupted users. Theoretical analyses establish that the regret performance of our proposed algorithm is better than the state-of-the-art online influence maximization algorithms. Extensive empirical evaluations on synthetic and real-world datasets also show the superior performance of our proposed algorithm. © 2013 IEEE.

关键词： Expectation maximization algorithm

来源：评论

学校读者我要写书评

暂无评论

InterACT: Enhancing Human-AI Interaction 2

InterACT: Enhancing Human-AI Interaction

引用

2nd International Conference on Machine Learning and Autonomous Systems, ICMLAS 2025

作者： Katta, Jithendra Kodumuru, Nikhil Reddy Katta, Swathi School of Engineering and Applied Sciences The George Washington University WashingtonDC United States Vignana Bharathi Institute of Technology Computer Science & Engineering dept. Hyderabad India

ISBN: (纸本)9798331505745

The rapid advancement of large language models and computer vision systems has opened new frontiers in artificial intelligence. This paper introduces InterACT, a novel cross-modal system that integrates leading language and vision models to enable more intuitive and context-aware human AI interactions. By leveraging the strengths of both modalities, InterACT processes natural language queries to guide object detection in visual environments, bridging the gap between linguistic and visual understanding. We evaluate the performance of three language models of varying sizes in conjunction with an advanced object detection model, exploring the balance between computational efficiency and interaction quality. Our experiments demonstrate InterACT's potential across diverse real-world scenarios, revealing promising results in maintaining contextual relevance and adaptability. This research contributes to the growing field of cross-modal AI systems and lays the groundwork for more sophisticated human AI interactions, with implications for applications ranging from assistive technologies to educational tools. © 2025 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

A Multimodal Monitoring System with XGBoost Classifier for Optimizing Student Engagement in Online Learning

A Multimodal Monitoring System with XGBoost Classifier for O...

引用

International Conference on Electrical, computer and Communication engineering (ECCE)

作者： Muhammad Kamal Hossen Mohammad Shorif Uddin Department of Computer Science and Engineering Chittagong University of Engineering and Technology Chattogram Bangladesh Department of Computer Science and Engineering Jahangirnagar University Dhaka Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Online learning has gained significant momentum due to globalization and the COVID-19 pandemic, but maintaining student engagement in this environment remains challenging. Traditional methods like self-reports and manual observations often miss subtle attentiveness cues, which necessitate more advanced tools for assessing engagement. This paper introduces a system that integrates computer vision techniques with an XGBoost classifier to analyze behaviors such as facial expressions, hand movements, mobile phone usage, and head poses via a standard webcam. The system outperformed traditional methods by achieving a 96.7% accuracy after validating on a dataset of 4,000 records. It generates detailed, anonymous reports, which offer current insights into student involvement while maintaining privacy. The system is resource-efficient and scalable to make it adaptable to various educational settings. A user-friendly interface enables seamless integration with e-learning platforms, allowing instructors to make data-driven interventions. This study also discusses the ethical and privacy concerns of such systems to balance the need for effective monitoring with the protection of student autonomy.

关键词： Privacy Webcams Pandemics Manuals Machine learning Mobile handsets Magnetic heads Protection Monitoring Standards

来源：评论

学校读者我要写书评

暂无评论

Integrating Machine Learning & Deep Learning for Image Captioning and Text Recognition in Mobile Apps Development

Integrating Machine Learning & Deep Learning for Image Capti...

引用

International Conference on Electrical, computer and Communication engineering (ECCE)

作者： J. Rahman S. R. Bhowmik Akib Jayed Islam Sultanus Salehin Muhammad Jawad Chowdhury Computer Science & Engineering Port City International University Chattogram Bangladesh Department of Computer Science Norwegian University of Science & Technology Gjøvik Norway Electrical & Computer Engineering The University of Alabama Tuscaloosa USA Computer Science & Engineering Islamic University of Technology Gazipur Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Deep learning methods are increasingly being applied to mobile applications for word recognition and image captioning, offering innovative solutions to complex tasks. In this study, models such as VGG19, ResNet101, and InceptionV3 Hybrid models, functioning as encoders in conjunction with LSTM decoders, are leveraged to generate natural language descriptions of images using the Flickr8k dataset. Furthermore, various machine learning techniques—such as Support Vector Machines (SVM), Random Forest, K-Nearest Neighbors (KNN), Logistic Regression, Naive Bayes, and Decision Trees—are evaluated for Bangla text categorization. Deep learning models, including CNNs, hybrid models, and VGG19, are applied for classifying English text and non-text images. A unique hybrid model combining CNN, VGG19, and LSTM is introduced to differentiate between text and non-text images in the English dataset. For nontext images, vision-based frameworks are employed to generate image captions. Moreover, two mobile applications developed using Android Studio are presented: the first enables live text scanning with text-to-speech conversion, while the second supports OCR-based text recognition from gallery images, providing voice outputs in both Bangla and English. Additionally, the research extends to handwritten word recognition through CNNs, RNNs, and Connectionist Temporal Classification (CTC) models, alongside CAPTCHA image recognition utilizing a combination of CNNs, OCR, and CTC techniques.

关键词： Deep learning Support vector machines Image recognition Text recognition Optical character recognition Text categorization Nearest neighbor methods Mobile applications Convolutional neural networks Long short term memory

来源：评论

学校读者我要写书评

暂无评论

CAST: Contrastive Analysis of Spatial and Temporal Features for QIM-Based VoIP Steganalysis

引用

IEEE Signal Processing Letters 2025年 32卷 1935-1939页

作者： Zhang, Cheng Yan, Yue Jiang, Shujuan Chen, Zhong China University of Mining and Technology School of Computer Science and Technology Engineering Research Center of Mine Digitalization Ministry of Education Xuzhou221116 China Swinburne College of Shandong University of Science and Technology Shandong 250031 China China University of Mining and Technology School of Information and Control Engineering Xuzhou221116 China

QIM(quantization index modulation)-based VoIP steganography is an information-hiding technology that malicious users could misuse to engage in illegal activities. Its countermeasure, commonly known as the QIM-based VoIP steganalysis, has been one of the research hotspots over the past decades. VoIP speech data is sequential, so most previous studies have focused on the temporal features extracted from VoIP encoding codewords to improve detection accuracy. As a result, spatial features are often ignored or fully investigated. In practice, VoIP speech data has unique spatial characteristics. Spatial features could capture the properties of nearby codewords and frames. Inspired by the success of CLIP (contrastive language–image pre-training), we propose a novel model that could efficiently incorporate spatial and temporal features named CAST (contrastive analysis of spatial and temporal features). CLIP introduces the concept of contrastive language–image learning, which has demonstrated exceptional efficiency and effectiveness in aligning textual and visual representations. However, it is not directly applicable to the field of QIM-based VoIP steganalysis. In CAST, we introduce a way to align spatial and temporal features through contrastive analysis. Experimental results demonstrate that this alignment is resource-efficient and could enhance detection accuracy. Meanwhile, CAST outperforms other state-of-the-art models in most scenarios. © 1994-2012 IEEE.

关键词： Encoding (symbols)

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：