检索结果-内蒙古大学图书馆

Meaningful image encryption algorithm based on compressive sensing and integer wavelet transform

Frontiers of computer science 2023年第3期17卷 161-175页

作者： Xiaoling HUANG Youxia DONG Guodong YE Yang SHI Faculty of Mathematics and Computer Science Guangdong Ocean UniversityZhanjiang 524088China School of Software Engineering Tongji UniversityShanghai 200092China

A new meaningful image encryption algorithm based on compressive sensing(CS)and integer wavelet transformation(IWT)is proposed in this *** of all,the initial values of chaotic system are encrypted by RSA algorithm,and then they are open as public *** make the chaotic sequence more random,a mathematical model is constructed to improve the random ***,the plain image is compressed and encrypted to obtain the secret ***,the secret image is inserted with numbers zero to extend its size same to the plain *** applying IWT to the carrier image and discrete wavelet transformation(DWT)to the inserted image,the secret image is embedded into the carrier ***,a meaningful carrier image embedded with secret plain image can be obtained by inverse ***,the measurement matrix is built by both chaotic system and Hadamard matrix,which not only retains the characteristics of Hadamard matrix,but also has the property of control and synchronization of chaotic ***,information entropy of the plain image is employed to produce the initial conditions of chaotic *** a result,the proposed algorithm can resist known-plaintext attack(KPA)and chosen-plaintext attack(CPA).By the help of asymmetric cipher algorithm RSA,no extra transmission is needed in the *** simulations show that the normalized correlation(NC)values between the host image and the cipher image are *** is to say,the proposed encryption algorithm is imperceptible and has good hiding effect.

关键词： image encryption algorithm compressive sensing integer wavelet transform Hadamard matrix

来源：评论

学校读者我要写书评

暂无评论

The Large Language Models on Biomedical Data Analysis: A Survey

引用

IEEE Journal of Biomedical and Health Informatics 2025年第6期29卷 4486-4497页

作者： Lan, Wei Tang, Zhentao Liu, Mingyang Chen, Qingfeng Peng, Wei Chen, Yi-Ping Phoebe Pan, Yi Guangxi University School of Computer Electronic and Information Nanning530004 China Kunming University of Science and Technology Faculty of Information Engineering and Automation Kunming650500 China La Trobe University Department of Computer Science and Information Technology MelbourneVIC3086 Australia Shenzhen University of Advanced Technology Faculty of Computer Science and Control Engineering Shenzhen518055 China

With the rapid development of Large Language Model (LLM) technology, it has become an indispensable force in biomedical data analysis research. However, biomedical researchers currently have limited knowledge about LLM. Therefore, there is an urgent need for a summary of LLM applications in biomedical data analysis. Herein, we propose this review by summarizing the latest research work on LLM in biomedicine. In this review, LLM techniques are first outlined. We then discuss biomedical datasets and frameworks for biomedical data analysis, followed by a detailed analysis of LLM applications in genomics, proteomics, transcriptomics, radiomics, single-cell analysis, medical texts and drug discovery. Finally, the challenges of LLM in biomedical data analysis are discussed. In summary, this review is intended for researchers interested in LLM technology and aims to help them understand and apply LLM in biomedical data analysis research. © 2013 IEEE.

关键词： Data consistency

来源：评论

学校读者我要写书评

暂无评论

Exploring Meta Evidence for Prompt Optimization

Exploring Meta Evidence for Prompt Optimization

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： He, Yao Liu, Jiajun Ke, Wenjun Wang, Peng Xiao, Congda Li, Yining School of Institute of Collaborate Innovation University of Macau China School of Computer Science and Engineering Southeast University Nanjing China

ISBN: (纸本)9798350368741

Prompt optimization aims to construct effective prompts to fully leverage the capabilities of large language models (LLMs). The key challenge lies in analyzing the outputs of LLMs, identifying fine-grained errors as explicit evidence and guiding reverse prompt correction. However, existing dominant methods feed all evidence into LLMs, leading to excessive redundancy that misleads the models. Moreover, this redundant input could increase computational costs. To address the issue, we introduce a novel meta evidence-aware prompt optimization method (MEPO), which performs fine-grained analysis of the causal relationships within evidence using clustering and Bayesian inference. Initially, MEPO generates preliminary evidence to describe the drawbacks of the given prompt. Subsequently, the evidence is clustered and constructed into a directed acyclic graph (DAG) to produce meta evidence. Finally, we edit the initial prompt based on the obtained meta evidence and employ the beam search to select the best-performing prompt. Experimental results show that MEPO outperforms state-of-the-art baselines, achieving absolute F1 score improvements of 5.8% and 5.0% on two common-used Liar and Jailbreak datasets, respectively. The datasets and codes are available at https://***/r/MEPO-E1A0. © 2025 IEEE.

关键词： Large Language Models Prompt Optimization

来源：评论

学校读者我要写书评

暂无评论

Dynamic Visual Semantic Sub-Embeddings and Fast Re-Ranking for Image-Text Retrieval

引用

IEEE Transactions on Multimedia 2025年 27卷 3781-3796页

作者： Wei, Wenzhang Gui, Zhipeng Wu, Changguang Zhao, Anqi Peng, Dehua Wu, Huayi Wuhan University School of Remote Sensing and Information Engineering Wuhan430079 China Wuhan University State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing Collaborative Innovation Center of Geospatial Technology Wuhan430079 China Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing210094 China

The core of image-text retrieval is to accurately measure the similarity between different modalities in a unified representation space. However, compared to textual descriptions of a certain perspective, the visual modality has more semantic variations. Therefore, images are usually associated with multiple textual captions in databases. Although popular symmetric embedding methods have explored numerous modal interaction approaches, they often learn toward outputting the average representation of multiple semantic variations within image embeddings. Consequently, information entropy in embeddings is increased, resulting in redundancy and decreased accuracy. In this work, we propose a Dynamic Visual Semantic SubEmbeddings framework (DVSE) to reduce the information entropy. Specifically, we obtain a set of heterogeneous visual sub-embeddings through dynamic orthogonal constraint loss. To encourage the generated candidate image embeddings to capture various semantic variations, we construct a mixed distribution and employ a variance-aware weighting loss to assign different weights to the optimization process. In addition, we develop a Fast Re-ranking strategy (FR) to efficiently evaluate the retrieval results and enhance the *** compare the performance with existing set-based method using five image feature encoders and three text feature encoders on three benchmark datasets: MSCOCO, Flickr30K and CUB Captions. We also show the role of different components by ablation studies and perform a sensitivity analysis of the hyperparameters. The qualitative analysis of visualized bidirectional retrieval and attention maps further demonstrates the ability of our method to encode semantic variations. © 1999-2012 IEEE.

关键词： Sensitivity analysis

来源：评论

学校读者我要写书评

暂无评论

Lag Synchronization and Lag H∞ Synchronization for Multiweighted Coupled Reaction-Diffusion Neural Networks Suffering Topology Attacks

引用

IEEE Transactions on Control of Network Systems 2025年第2期12卷 1488-1499页

作者： Wang, Jin-Liang Zhao, Xin-Yu Ren, Shun-Yan Huang, Tingwen Tiangong University Tianjin Key Laboratory of Autonomous Intelligence Technology and Systems School of Computer Science and Technology Tianjin300387 China Tiangong University School of Electronics & Information Engineering Tianjin300387 China University of Electronic Science and Technology of China School of Mechanical and Electrical Engineering Chengdu611731 China Shenzhen University of Advanced Technology Faculty of Computer Science and Control Engineering Shenzhen518055 China

This paper mainly discusses two kinds of coupled reaction-diffusion neural networks (CRNN) under topology attacks, that is, the cases with multistate couplings and with multiple spatial-diffusion couplings. On one hand, by adopting state feedback control scheme, the problem of the lag synchronization for the multistate coupled CRNN is addressed. Moreover, the influences of external disturbances for the lag synchronization of the CRNN with multistate couplings are also taken into consideration. On the other hand, several lag and lag H∞ synchronization conditions for CRNN with multiple spatial-diffusion couplings are derived under the help of the Kronecker product and the state feedback control method. Finally, two numerical examples are provided to verify the correctness of the derived criteria. © 2014 IEEE.

关键词： State feedback

来源：评论

学校读者我要写书评

暂无评论

Comprehensive multimodal approach for Parkinson’s disease classification using artificial intelligence: insights and model explainability

引用

Soft Computing 2025年第3期29卷 1845-1877页

作者： Balaha, Hossam Magdy Hassan, Asmaa El-Sayed Ahmed, Rawan Ayman Balaha, Magdy Hassan Bioengineering Department J.B. Speed School of Engineering University of Louisville LouisvilleKY United States Computer Science and Control Systems Engineering Department Faculty of Engineering Mansoura University Mansoura Egypt Mathematics and Engineering Physics Department Faculty of Engineering Mansoura University Mansoura Egypt Department of Obstetrics and Gynecology Faculty of Medicine Tanta University Tanta Egypt

Parkinson’s disease (PD) is a debilitating neurodegenerative disorder affecting millions worldwide. Early detection is vital for effective management, yet remains challenging. In this study, we investigated four distinct datasets for PD detection. Through comprehensive experimentation employing ensemble methods and feature selection, we achieved high classification accuracies across the datasets. For the Oxford Parkinson’s Disease Detection Dataset, an accuracy of 95.67%, precision of 97.59%, recall of 84.5%, specificity of 99.32%, and F1-score of 90.57% were achieved. For the Alzheimer Parkinson Diseases 3 Class Dataset, the "Stacking" approach surpasses individual models, reaching an accuracy of 99.85%, precision of 99.81%, recall of 99.81%, specificity of 99.86%, and F1 of 99.81%. For the NewHandPD dataset, Regarding the Spiral category, The "Base-P32-384" model surpasses others with an accuracy of 97.35%, precision of 96.50%, recall of 98.57%, and F1-score of 97.53%. The collective "Stacking" approach proves highly effective regarding the Circle category, achieving 100% across all performance metrics. Regarding the Meander category, the "Base-P16-224" model achieves an accuracy of 97.35%, precision of 99.26%, recall of 95.71%, specificity of 99.19%, and F1 of 97.45%. The Mobile Device Voice Recordings at King’s College London (MDVR-KCL) dataset contains two datasets. Regarding the "SpontaneousDialogue" dataset, accuracy, BAC, precision, recall, specificity, and F1-score were computed, resulting in values of 94.03%, 92.83%, 90.78%, 100.0%, and 85.67%, respectively. Regarding the "ReadText" dataset, accuracy, BAC, precision, recall, specificity, and F1-score were computed, resulting in values of 91.89%, 90.62%, 87.5%, 100.0%, and 81.25%, respectively. Our findings highlight the efficacy of leveraging diverse data sources and advanced machine learning techniques to enhance PD detection accuracy. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germa

关键词： Neurodegenerative diseases

来源：评论

学校读者我要写书评

暂无评论

AI-Driven Learning Management Systems:Modern Developments, Challenges and Future Trends during theAge of ChatGPT

引用

computers, Materials & Continua 2024年第8期80卷 3289-3314页

作者： Sameer Qazi Muhammad Bilal Kadri Muhammad Naveed Bilal AKhawaja Sohaib Zia Khan Muhammad Mansoor Alam Mazliham Mohd Su’ud Department of Computer Science Faculty of EngineeringScience and TechnologyIQRA UniversityKarachi75500Pakistan College of Computer and Information Sciences Prince Sultan UniversityRiyadh11586Saudi Arabia Faculty of Engineering Department of Electrical EngineeringIslamic University of MadinahP.O.BOX 170Madinah41411Saudi Arabia Faculty of Engineering Department of Mechanical EngineeringIslamic University of MadinahP.O.BOX 170Madinah41411Saudi Arabia Faculty of Computing Riphah International UniversityIslamabad46000Pakistan Faculty of Computing and Informatics(FCI)Multimedia University Cyberjaya63100Malaysia Faculty of Engineering and Information Technology School of Computer ScienceUniversity of Technology SydneySydneyNSW 2007Australia

COVID-19 pandemic restrictions limited all social activities to curtail the spread of the *** foremost and most prime sector among those affected were schools,colleges,and *** education system of entire nations had shifted to online education during this *** shortcomings of Learning Management Systems(LMSs)were detected to support education in an online mode that spawned the research in Artificial Intelligence(AI)based tools that are being developed by the research community to improve the effectiveness of *** paper presents a detailed survey of the different enhancements to LMSs,which are led by key advances in the area of AI to enhance the real-time and non-real-time user *** AI-based enhancements proposed to the LMSs start from the Application layer and Presentation layer in the form of flipped classroom models for the efficient learning environment and appropriately designed UI/UX for efficient utilization of LMS utilities and resources,including AI-based *** layer enhancements are also required,such as AI-based online proctoring and user authentication using *** extend to the Transport layer to support real-time and rate adaptive encrypted video transmission for user security/privacy and satisfactory working of *** also needs the support of the Networking layer for IP-based geolocation features,the Virtual Private Network(VPN)feature,and the support of Software-Defined Networks(SDN)for optimum Quality of Service(QoS).Finally,in addition to these,non-real-time user experience is enhanced by other AI-based enhancements such as Plagiarism detection algorithms and Data Analytics.

关键词： Learning management systems chatbots ChatGPT online education Internet of Things(IoT) artificial intelligence(AI) convolutional neural networks natural language processing

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Lung Cancer Histopathology Classification: Emphasizing Accuracy and Customisation

Biomedical Materials & Devices

引用

Biomedical Materials & Devices 2025年 1-13页

作者： Kashyap, Seema Shukla, Arvind Kumar Naim, Iram Pal, Saurabh School of Computer Science and Engineering IFTM University Moradabad India Faculty of Engineering and Technology MJP Rohilkhand University Bareilly India Faculty of Engineering and Technology VBS Purvanchal University Jaunpur India

Lung cancer is a dangerous disease that can be fatal, and a correct diagnosis is essential for figuring out the best way to treat it. The optimum treatment for people with lung cancer requires the classification of the disease into its histological types, such as adenocarcinoma (ADC), small cell lung cancer (SCLC), and squamous cell carcinoma (SCC). Each histological subtype has its features and may react differently to different types of medicine. So, knowing the exact subtype helps guide treatment choices and improve the patient's outcome. Lung cancer subtypes are necessary for personalized treatment. It helps doctors choose tumor-specific treatments such as surgery, radiation, chemotherapy, targeted drugs, and immunotherapies. Precise categorization improves prognosis, avoids needless medicines, and lets patients participate in clinical studies targeting their cancer subtype. Precision medicine improves lung cancer outcomes with accurate categorization. The current algorithms in this domain have shown deficiencies in performance criteria such as specificity, F-score, sensitivity, and precision in recognition. These limitations may stem from challenges such as the complexity and heterogeneity of histopathological images, variations in staining techniques, and the presence of confounding factors. Deep learning methods have made it easier to look at histopathology slides of cancer and see what's going on. Several studies have shown that convolutional neural networks (CNN) are essential for classifying histopathological pictures of different kinds of cancer, like brain, skin, breast, lung, and colon cancer. This study divides lung cancer images into three groups: normal, adenocarcinoma, and squamous cell carcinoma. We have been training deep learning algorithms to identify lung cancer in histopathology slides better, and utilizing deep learning strategies and cutting-edge algorithms such as VGG-19, ResNet-50 v2, EfficientNetB1, and others indicates a comprehensive ap

关键词： And images from histopathology Classification Deep learning EfficientNetB1 Lung cancer ResNet-50 v2 VGG-19

来源：评论

学校读者我要写书评

暂无评论

Pager Explosion:Cybersecurity Insights and Afterthoughts

引用

IEEE/CAA Journal of Automatica Sinica 2024年第12期11卷 2359-2362页

作者： Chuan Sheng Wanlun Ma Qing-Long Han Wei Zhou Xiaogang Zhu Sheng Wen Yang Xiang Fei-Yue Wang Swinburne University of Technology Hawthorn VIC 3122Australia IEEE the School of Computer and Mathematical Sciences the University of AdelaideAdelaideSouth Australia 5005Australia the State Key Laboratory for Management and Control of Complex Systems Chinese Academy of SciencesBeijing 100190 the Department of Engineering Science Faculty of Innovation EngineeringMacao University of Science and TechnologyMacao 999078China

PAGER EXPLOSION:THE *** 2024 Lebanon pager explosions represent one of the most unexpected and devastating technological incidents in recent *** September 17 and 18,2024,thousands of pagers and walkie-talkies exploded simultaneously across Lebanon and parts of Syria,resulting in 42 deaths and more than 3500 *** handheld communication devices,previously regarded as secure and low-profile,were rigged with concealed explosives and remotely triggered by attackers.

关键词： injuries thoughts explosion

来源：评论

学校读者我要写书评

暂无评论

Remote Sensing Classification of River Using Machine Learning Methods 12

Remote Sensing Classification of River Using Machine Learnin...

引用

12th Academic Conference of Geology Resource Management and Sustainable Development, GRMSD 2024

作者： Cui, Hanwen Li, Cheng Tian, Xiaolin School of Computer Science Zhuhai College of Science and Technology Guangdong Zhuhai519041 China Academy of Interdisciplinary Studies Hong Kong University of Science and Technology 999077 Hong Kong School of Computer Science and Engineering Faculty of Innovation Engineering Macau University of Science and Technology 999078 China

ISBN: (纸本)9798331314545

In the realm of global sustainable development, effective management of water resources and the protection of ecological environments have increasingly emerged as focal points for both governmental and international organizations. Rivers, as critical conduits of water resources, fulfill various essential functions, including transportation, irrigation, aquaculture, tourism, and climate regulation, in addition to their provision of water for human consumption and agricultural purposes. This study predominantly explores the application of remote sensing technology and machine learning methodologies to classify rivers. The research is situated in Zhuhai City, characterized by a vertically distributed river network. It employs eXtreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM), both based on gradient boosting decision trees (GBDT). The data source for this investigation consists of high spatial resolution imagery from Gaofen-2 (GF-2) remote sensing satellites. The results indicate that both models achieve an overall accuracy (OA) exceeding 96.50%, with Kappa coefficients surpassing 0.92. Notably, the LightGBM model demonstrates superior performance over XGBoost in terms of both classification accuracy and operational efficiency. The utilization of remote sensing images in conjunction with machine learning techniques for the classification and monitoring of rivers offers a robust scientific framework for enhancing water resource management. Furthermore, these methodologies can assist in the formulation of effective strategies for the sustainable utilization and protection of water resources. © 2024 Proceedings of the 12th Academic Conference of Geology Resource Management and Sustainable Development, GRMSD 2024. All rights reserved.

关键词： Rivers

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：