检索结果-内蒙古大学图书馆

CMV2U-Net: A U-shaped network with edge-weighted features for detecting and localizing image splicing

Journal of Forensic sciences 2025年第0期70卷 2025 Apr 3页

作者： Akram, Arslan Jaffar, Muhammad Arfan Rashid, Javed Boulaaras, Salah Mahmoud Faheem, Muhammad Faculty of Computer Science and Information Technology The Superior University Lahore Pakistan Information Technology Services University of Okara Okara Pakistan Department of Mathematics College of Science Qassim University Buraydah Saudi Arabia VTT Technical Research Centre of Finland Espoo Finland

The practice of cutting and pasting portions of one image into another, known as “image splicing,” is commonplace in the field of image manipulation. Image splicing detection using deep learning has been a hot research topic for the past few years. However, there are two problems with the way deep learning is currently implemented: first, it is not good enough for feature fusion, and second, it uses only simple models for feature extraction and encoding, which makes the models vulnerable to overfitting. To tackle these problems, this research proposes CMV2U-Net, an edge-weighted U-shaped network-based image splicing forgery localization approach. An initial step is the development of a feature extraction module that can process two streams of input images simultaneously, allowing for the simultaneous extraction of semantically connected and semantically agnostic features. One characteristic is that a hierarchical fusion approach has been devised to prevent data loss in shallow features that are either semantically related or semantically irrelevant. This approach implements a channel attention mechanism to monitor manipulation trajectories involving multiple levels. Extensive trials on numerous public datasets prove that CMV2U-Net provides high AUC and F₁ in localizing tampered regions, outperforming state-of-the-art techniques. Noise, Gaussian blur, and JPEG compression are post-processing threats that CMV2U-Net has successfully resisted. © 2025 American Academy of Forensic sciences.

关键词： digital forensics feature fusion image authentication image forgery localization image splicing detection U-shaped network

来源：评论

学校读者我要写书评

暂无评论

Cocktail Party Effect Using Parallel Intra and Inter Self-attention 3rd

Cocktail Party Effect Using Parallel Intra and Inter Self-at...

引用

3rd International Conference on Advances in Information and Communication technology, ICTA 2024

作者： Tan, Ha Minh Quoc, Nguyen Kim Vu, Duc-Quang Dang, An Thi, Duyen Nguyen Faculty of Information Technology Nguyen Tat Thanh University Thai Nguyen Viet Nam Department of Computer Science and Information System Thai Nguyen University of Education Thai Nguyen Viet Nam Faculty of Computer Science Phenikaa University Ha Noi Viet Nam Thai Nguyen University of Information and Communication Technology Thai Nguyen Viet Nam

ISBN: (纸本)9783031809422

The self-attention architecture is also considered to be a significant contribution in sequence processing tasks. It has the ability to highlight the distinctive features of a sequence, has been very successful in natural language processing and speech separation. In addition, the intra- and inter-architecture of the dual-path bi-directional long-short term memory framework has been demonstrated to be very effective for processing extremely long sequences. Based on these properties, we design a core architecture called intra- and inter-self-attention framework. Parallel models is more efficient and stable than a single model. In this research, we propose a parallel intra- and inter-self-attention framework (PI2SA) for the cocktail party effect task. The experimental results demonstrate that our PI2SA is remarkably effective. It outperforms both traditional self-attention models and a basic intra- and inter-architecture. Our PI2SA achieves competitive results with state-of-the-art deep learning models on the cocktail party effect problem. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Ontology-Driven Semantic Web Framework with Sentiment Analysis for Risk Assessment and Diagnosis of Diabetes Mellitus 4

Ontology-Driven Semantic Web Framework with Sentiment Analys...

引用

4th International Conference on Sentiment Analysis and Deep Learning, ICSADL 2025

作者： Muruganantham, A. Velmurugan, R. Kumar, R. Rajan, M. Sundar Sivaraj, V. Karnataka Bengaluru India Arba Minch Institute of Technology Arba Minch University Faculty of Electrical and Computer Engineering Ethiopia Karpagam Academy of Higher Education Department of Computer Science Coimbatore India

ISBN: (纸本)9798331523923

This paper uses semantic web and ontology techniques to predict the risk analysis of patients with diabetes mellitus. The data is collected from patients through personal interaction and by accessing their previous medical history. However, during personal interactions, medical staff often fail to gather all the necessary details for an accurate diagnosis, and patients may lose their previous medical records, which hinders doctors' ability to gain a full understanding of the problem. To address this issue, ontology techniques are employed, and additional details are collected from patients, such as their food habits, smoking history, alcohol consumption, sexual activity, and cardiovascular health. Using the collected data, ontology and semantic web technologies are applied to predict the risk analysis of diabetic patients. The results are visualized in a graph, providing a more comprehensive and accurate assessment of the patients' health risks. © 2025 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Optimizing Public Health: A Nationwide Vaccine Distribution Management System 11th

Optimizing Public Health: A Nationwide Vaccine Distribution ...

引用

11th International Symposium on Integrated Uncertainty in Knowledge Modelling and Decision Making, IUKM 2025

作者： Abbas, Maythem K. Alashhab, Abdussalam Ahmed Majeed, Samer Hameed Rashid, Omar Fitian Asia Pacific University of Technology and Innovation Technology Park Malaysia Bukit Jalil Kuala Lumpur57000 Malaysia Department of Computer Science Faculty of Information Technology Alasmarya Islamic University Zliten Libya Longo Faculty of Business/Humber College TorontoON Canada College of Science University of Baghdad Baghdad Iraq

ISBN: (纸本)9789819646029

During the last hit of Coronavirus disease (COVID-19), all the countries were troubled by as they were not ready for, thus many serious consequences had popped up to the social, economic, and health aspects of our lives. Hence, people should be vaccinated as soon as possible to eradicate the pandemic. The countries faced difficulties those, generally, revolve around tracking, queue management, privacy, data security, accessibility, and other operational challenges. This project is concerned about solving the above problems by proposing a mobile application for nationwide vaccine distribution management. Three similar existing systems, namely MySejahtera, COVID-19 UAE, and Halton Region were reviewed and analyzed based on their features and weaknesses to get the requirements of the proposed solution. The important features included in the proposed solution are register and manage vaccination appointments, book vaccination for dependents, risk assessment tool, pandemic statistics, and announcements. Apart from that, the evolutionary prototyping model was used in this project to develop the proposed solution. By the deployment of such an application, the world will be ready to face any future pandemic without being troubled. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Lightweight Attention-CycleGAN for Nighttime-Daytime Image Transformation 1

引用

2nd International Conference on Artificial Intelligence Security and Privacy, AIS and P 2024

作者： Huang, Junhao Xiao, Xiangjun Zhou, Haojun Yasin, Affan Zhou, Zhili School of Computer Science and Cyber Engineering Guangzhou University Guangzhou China Faculty of Science and Technology Universitas Airlangga Surabaya Indonesia School of Artificial Intelligence Guangzhou University Guangzhou China

ISBN: (数字)9789819611485

ISBN: (纸本)9789819611478

With the rapid development of deep learning in the field of computer vision, the performance of core vision tasks such as image recognition has achieved significant improvement. In nighttime environment, due to the low-light condition and reduced visibility, cross-domain transformation of nighttime images based on Generative Adversarial Network (GAN) model can effectively improve the accuracy of nighttime recognition models. However, the existing GAN models are difficult to be effectively deployed on resource-constrained devices due to the requirement of high storage space and computational resource. To this end, this paper proposes a shared attention network based on the attention mechanism with the CycleGAN structure, and designs an online knowledge distillation method to compress and optimize the model, so as to obtain a lightweight model to achieve the nighttime-daytime cross-domain image transformation. Experimental results demonstrate that the proposed model achieves the state-of-the-art performance in the task of Nighttime-Daytime Image Transformation. This is of great significance for edge devices to solve the problem of recognition at night. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Hybrid Parallel Architecture Integrating FFN, 1D CNN, and LSTM for Predicting Wildfire Occurrences in Morocco 9th

Hybrid Parallel Architecture Integrating FFN, 1D CNN, and L...

引用

9th International Conference on Smart City Applications, SCA 2024

作者： Jadouli, Ayoub El Amrani, Chaker Computer Science and Smart Systems Faculty of Sciences and Technology Abdelmalek Essaâdi University Tangier Morocco

ISBN: (纸本)9783031886522

A novel “Hybrid Parallel Architecture Integrating FFN, 1D CNN, and LSTM” is presented to enhance wildfire prediction capabilities in Morocco. Utilizing the “Morocco Wildfire Predictions: 2010–2022 ML Dataset”, this model merges various data points, including meteorological conditions, human population density, and environmental factors like soil moisture and vegetation indices. The architecture combines feedforward neural networks (FFN), one-dimensional convolutional neural networks (1D CNN), and long short-term memory networks (LSTM) to process different subsets of features in parallel, capturing both spatial and temporal dependencies crucial for accurate wildfire predictions. The dataset covers data from 2010 to 2022 and includes standardized features specifically prepared for machine learning applications in disaster management. By training this hybrid model, significant validation accuracy was observed, indicating robust performance. Initial results were promising, with the model achieving an accuracy of 83.96% on the training set and 87.19% on the validation set during the first epoch, improving to 87.13% on the training set and 87.56% on the validation set by the second epoch. These results underscore the efficacy of integrating multiple neural network architectures to enhance predictive accuracy in critical applications such as wildfire prediction. The findings suggest substantial potential for deploying such models in real-world scenarios, aiding policymakers and disaster management teams in deploying timely and effective responses to mitigate wildfire impacts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Disaster Management Environmental Factors Feedforward Neural Network Hybrid Parallel Architecture Long Short-Term Memory Machine Learning Meteorological Data Morocco One-Dimensional Convolutional Neural Network Wildfire Prediction

来源：评论

学校读者我要写书评

暂无评论

Enhanced English Sentence Pair Classification Through KIRINLog Technique 5

Enhanced English Sentence Pair Classification Through KIRINL...

引用

5th International Conference on Advances in Electrical, Electronics and Computing technology, EECT 2025

作者： Ounprasert, Patthanan Anekboon, Khantharat Department of Computer and Information Sciences Faculty of Applied Science King Mongkut's University of Technology North Bangkok Bangkok Thailand

ISBN: (纸本)9798331541545

Natural language processing is a rapidly evolving technology in an era of quick technological growth. Conventional rule-based methods are insufficient to meet the demands of today's complex challenges. One of the significant tasks in natural language processing is recognizing textual entailment. Recognizing textual entailment is vital in natural language processing as it enables systems to understand the relationship between texts, which is crucial for applications like text summarization, sentiment analysis, information verification, question answering, text classification, and machine translation. One challenge is reducing the size of the input vectors while maintaining good prediction results, F1 scores. To overcome this challenge, this research presents the novel encoding technique by reducing the encoding size for the sentences involving compositional knowledge dataset. This dataset is sentence pairs in English to classify textual entailments into three categories: entailment, neutral, and contradiction. This research proposes the KIRINLog technique, which consists of six main methods: preprocessing, removing the same word, max-length padding and truncation for sentence alignment, adding a flag value, word embedding by Word2Vec, and dimensionality re-duction with principal component analysis to reduce the size of input vector. All of six main processes are proposed to create feature vectors for recognizing textual entailment (the sentences involving compositional knowledge dataset). This study used an attention-based bidirectional LSTM as a classification model. This model combines bidirectional LSTM with an attention to make analyzing sequences in both forward and backward directions and to help the model focus on important information in the sentence. This proposed technique achieved an accuracy of 92.4%, with precision at 88.8%, recall at 88.6%, and an F1-score of 88.6% on the 2014 sentences involving compositional knowledge dataset. © 2025 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

A novel U-net model for brain tumor segmentation from MRI images

引用

Alexandria Engineering Journal 2025年 126卷 220-230页

作者： Obayya, Marwa Alshuhail, Asma Mahmood, Khalid Alanazi, Meshari H. Alqahtani, Mohammed Aljehane, Nojood O. Almansour, Hamad Al-Hagery, Mohammed Abdullah Department of Biomedical Engineering College of Engineering Princess Nourah bint Abdulrahman University PO Box 84428 Riyadh11671 Saudi Arabia Department of Information Systems College of Computer Sciences & Information Technology King Faisal university Saudi Arabia Department of Information Systems Applied College at Mahayil King Khalid University Saudi Arabia Department of Computer Science College of Sciences Northern Border University Arar Saudi Arabia Department of Information System and Cyber Security College of Computing and Information Technology University of Bisha Bisha61922 Saudi Arabia Department of Computer Science Faculty of Computers and Information Technology University of Tabuk Tabuk Saudi Arabia Applied Collage Najran University Najran Saudi Arabia Department of Computer Science College of Computer Qassim University Saudi Arabia

Segmentation of brain tumors aids in diagnosing the disease early, planning treatment, and monitoring its progression in medical image analysis. Automation is necessary to eliminate the time and variability associated with traditional segmentation methods. Convolutional neural networks (CNNs) and U-Net architectures have demonstrated their efficiency and effectiveness in segmenting brain tumors from MRI images using deep learning techniques. The paper presents an improved U-Net-based segmentation algorithm that integrates nested skip paths to improve encoder-decoder feature fusion. The performance of segmentation was optimized by utilizing a variety of activation functions and loss functions, including Dice Loss and Intersection over Union (IoU). A high level of accuracy was demonstrated in the proposed model when it was evaluated using the LGG Segmentation Dataset. The proposed approach for segmenting medical images has been shown to be both robust and efficient in a comparative analysis. © 2025 The Authors

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Identifying Neural Biomarkers of Confirmation Bias 13

Identifying Neural Biomarkers of Confirmation Bias

引用

13th International Winter Conference on Brain-computer Interface, BCI 2025

作者： Singh, Avinash Liu, Jia Voinov, Alexey Lin, Chin-Teng School of Computer Science University of Technology Sydney UltimoNSW Australia School of Biomedical Engineering University of Sydney CamperdownNSW Australia Faculty of Engineering Technology University of Twente Netherlands

ISBN: (纸本)9798331521929

It has been observed that individuals prefer to favour information that strengthens their beliefs or values, while information that contradicts them causes unpleasant feelings and is more likely to be ignored. This is known as confirmation bias, and it is usually difficult to dislodge once affirmed. Confirmation bias has been shown to lead to flawed decisions in political, organizational, financial and scientific domains. To better understand the origins and detection of neural biomarkers for confirmation bias, this article presents a study that employed electroencephalography (EEG) signals from 10 participants. The study revealed an apparent increase in theta activity (phase-resetting) and prediction error negativity (PEN) among participants exhibiting confirmation bias. These findings have the potential to contribute to the development of novel brain-computer interface systems capable of detecting confirmation bias. Such systems may help mitigate social polarization and the negative effects of biased decision-making. © 2025 IEEE.

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

Energy Harvesting Jammer-Aided Covert Communications in Wireless Multi-Relay IoT Systems

引用

IEEE Internet of Things Journal 2025年第11期12卷 17443-17455页

作者： Lv, Hao Yang, Bin Sun, Xiuwen Gao, Chan Gui, Bao Taleb, Tarik School of Computer Science and Technology of Anhui University Hefei230000 China Chuzhou University School of Computer and Information Engineering Chuzhou239000 China Xi'an University of Posts and Telecommunications National Engineering Research Center for Secured Wireless School of Cybersecurity Xi'an710121 China Ruhr University Bochum Faculty of Electrical Engineering and Information Technology Bochum44801 Germany

This paper investigates covert communications in a multi-relay Internet of Things (IoT) system with multiple energy harvesting jammers, where a transmitter (Alice) attempts to covertly transmit confidential messages to its destination (Bob) through relay forwarding, while a warden (Willie) detects the existence of Alice's transmission. Specifically, we employ a harvesting-then-jamming protocol with which the jammers first harvest energy from Alice and then send jamming signals to interfere with Willie's detection. We propose a relay and jammer selection strategy, namely quality of service (QoS)-aware selection, and use the random selection strategy as a comparison strategy. Under these two selection strategies, we derive the optimal detection threshold and minimum detection error probability at Willie, respectively. We then model the covert throughput performance and obtain the maximum covert throughput by jointly optimizing covert transmit power and jamming transmit power. Extensive numerical results are provided to illustrate the impacts of system parameters on covert throughput performance. © 2014 IEEE.

关键词： Jamming

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：