检索结果-内蒙古大学图书馆

RDDM: A Rate-Distortion Guided Diffusion Model for Leaned Image Compression Enhancement

IEEE Journal on Emerging and Selected Topics in Circuits and systems 2025年

作者： Jiang, Sanxin Katto, Jiro Sun, Heming Shanghai University of Electric Power Department of Information Engineering Shanghai China Waseda University Department of Computer Science and Communication Engineering Tokyo Japan Yokohama National University Faculty of Engineering Yokohama Japan

Currently, denoising diffusion probability models (DDPM) have achieved significant success in various image generation tasks, but their application in image compression, especially in the context of learned image compression (LIC), is quite limited. In this study, we introduce a rate-distortion (RD) guided diffusion model, referred to as RDDM, to enhance the performance of LIC. In RDDM, LIC is treated as a lossy codec function constrained by RD, dividing the input image into two parts through encoding and decoding operations: the reconstructed image and the residual image. The construction of RDDM is primarily based on two points. First, RDDM treats diffusion models as repositories of image structures and textures, built using extensive real-world datasets. Under the guidance of RD constraints, it extracts and utilizes the necessary structural and textural priors from these repositories to restore the input image. Second, RDDM employs a Bayesian network to progressively infer the input image based on the reconstructed image and its codec function. Additionally, our research reveals that RDDM’s performance declines when its codec function does not match the reconstructed image. However, using the highest bitrate codec function minimizes this performance drop. The resulting model is referred to as RDDM*. The experimental results indicate that both RDDM and RDDM* can be applied to various architectures of LICs, such as CNN, Transformer, and their hybrid. They can significantly improve the fidelity of these codecs while maintaining or even enhancing perceptual quality to some extent. © 2011 IEEE.

关键词： Image denoising

来源：评论

学校读者我要写书评

暂无评论

GLC++: Source-Free Universal Domain Adaptation through Global-Local Clustering and Contrastive Affinity Learning

arXiv

引用

arXiv 2024年

作者： Qu, Sanqing Zou, Tianpei Röhrbein, Florian Lu, Cewu Chen, Guang Tao, Dacheng Jiang, Changjun The School of Automotive Engineering Department of Computer Science Tongji University China The Faculty of Computer Science Chemnitz University of Technology Germany The Department of Computer Science Shanghai Jiao Tong University China The School of Computer Science Faculty of Engineering The University of Sydney Australia The Department of Computer Science Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University China

Deep neural networks often exhibit sub-optimal performance under covariate and category shifts. Source-Free Domain Adaptation (SFDA) presents a promising solution to this dilemma, yet most SFDA approaches are restricted to closed-set scenarios. In this paper, we explore Source-Free Universal Domain Adaptation (SF-UniDA) aiming to accurately classify "known" data belonging to common categories and segregate them from target-private "unknown" data. We propose a novel Global and Local Clustering (GLC) technique, which comprises an adaptive one-vs-all global clustering algorithm to discern between target classes, complemented by a local k-NN clustering strategy to mitigate negative transfer. Despite the effectiveness, the inherent closed-set source architecture leads to uniform treatment of "unknown" data, impeding the identification of distinct "unknown" categories. To address this, we evolve GLC to GLC++, integrating a contrastive affinity learning strategy. We examine the superiority of GLC and GLC++ across multiple benchmarks and category shift scenarios. Remarkably, in the most challenging open-partial-set scenarios, GLC and GLC++ surpass GATE by 16.7% and 18.6% in H-score on VisDA, respectively. GLC++ enhances the novel category clustering accuracy of GLC by 4.3% in open-set scenarios on Office-Home. Furthermore, the introduced contrastive learning strategy not only enhances GLC but also significantly facilitates existing methodologies. Copyright © 2024, The Authors. All rights reserved.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Advances in Finite Element Analysis for Cancer Therapy Focusing on Magnetic Nanoparticle Hyperthermia

引用

Multiscale science and engineering 2024年第2期6卷 113-123页

作者： Raouf, Izaz Gas, Piotr Kim, Heung Soo Department of Mechanical Robotics and Energy Engineering Dongguk University-Seoul Seoul Republic of Korea Department of Electrical and Power Engineering Faculty of Electrical Engineering Automatics Computer Science and Biomedical Engineering AGH University of Krakow Krakow Poland

This mini review provides a thorough examination of recent advances in finite element analysis (FEA) for cancer therapy, with a focus on magnetic nanoparticle (MNP) hyperthermia. The paper begins by discussing the fundamental principles of FEA and their applications in biomedical sciences. It then describes the application of FEA to the in vitro and in vivo modeling of MNP hyperthermia, including the development and validation of computational models that simulate magnetic nanoparticles behavior under alternating magnetic field (AMF), heat generation, its transfer, and cellular/tissue responses. The review further explores the emerging trends and future directions in the application of FEA for MNP hyperthermia, including the integration of experimental data, incorporation of patient-specific parameters, and optimization of treatment protocols. The paper also discusses the key challenges and limitations of current FEA models, shedding light on potential areas for future research and development. By synthesizing the most recent advances in this field, the review aims to provide a valuable resource for researchers, clinicians, and engineers working on the optimization and clinical translation of MNP hyperthermia for effective cancer therapy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Reassessing Glove Embeddings in Deep Learning: A Comparative Study with Classical ML Approaches

引用

Procedia computer science 2024年 251卷 740-745页

作者： Zakia Labd Said Bahassine Khalid Housni Fatima Zahrae Aithamou Aadi Laboratory of research in informatics L@RI Department of Computer Science Faculty of Sciences Ibn Tofail University Kenitra Morocco Laboratory of Artificial Intelligence and Complex Systems Engineering Department of Computer Science National Higher School of Arts and Crafts Hassan II University Casablanca Morocco

The importance of text classification algorithms has increased due to the growing availability of large-scale data. This has led to a greater demand for efficient classification techniques and encoding algorithms. Word embedding techniques, like Glove, have shown significant success in encoding semantic relationships between words. This research paper aims to reassess the effectiveness of Glove embeddings coupled with deep learning algorithms. The impact of Glove embedding on two widely used deep learning models: Recurrent Neural Networks (RNN) and Recurrent Convolutional Neural Networks (RCNN) is analyzed. The results highlight the impact of Glove embeddings on deep learning models, showcasing significant performance enhancements in some cases while having minimal effects in others. By examining the impact of Glove embeddings on traditional ML algorithms in a previous study, valuable context for understanding the performance differences between the two approaches is obtained.

关键词： Glove Text Classification, Deep Learning, Natural Language Processing (NLP), Recurrent Convolutional Neural Networks (RCNN)

来源：评论

学校读者我要写书评

暂无评论

Relationship of IQ with Causal Connections Between Brain Salient Network Regions

Relationship of IQ with Causal Connections Between Brain Sal...

引用

Iranian Conference of Biomedical engineering (ICBME)

作者： Mina Ghoorchian Hamid Soltanian-Zadeh Mir Mohsen Pedram Mohsen Mir Mohammad Sadeghi Department of Cognitive Science Institute for Cognitive Science Studies Tehran Iran School of Electrical and Computer Engineering University of Tehran Tehran Iran Department of Electrical and Computer Engineering Faculty of Engineering Kharazmi University Tehran Iran

ISBN: (数字)9798331529710

ISBN: (纸本)9798331529727

IQ is one of the indicators that has always been of interest to psychiatrists, doctors and cognitive science researchers. Since this index plays a key role in people’s lives and also in the occurrence of brain abnormalities, many studies have been conducted on it. In this paper, we used the resting state magnetic resonance imaging data available in the HCP database to classify people into three groups with low, medium, and high IQ and applied machine learning methods on the matrix of causal connections between related network regions. We applied brain highlights. We calculated the matrix of causal relationships in this network by the spDCM algorithm and then obtained the differences of the three groups using the ANOVA statistical test and 6 edges out of 169 edges with p-value <0.05 were found to be significantly different, according to the order of the edges, this difference is in the relationship between rMCC-rvIPFC, rInsula-rvIPFC, rInsula-rPutamen, rInsulaIIPG, IIPG-rSFG, IIPG-rSFG. In the next step, we used the algorithms of support machines and nearest neighbors to classify these three groups. Finally, using the support vector regression algorithm, we predicted the IQ from the edge values with RMSE=5.41 and MAE=3.59. The results of this research indicate a significant relationship between IQ and the salient network of the brain.

关键词： Support vector machines Neuroimaging Visualization Accuracy Machine learning algorithms Medical treatment Prediction algorithms Classification algorithms Indexes Analysis of variance

来源：评论

学校读者我要写书评

暂无评论

Combining Global and Local Merges in Logic-based Entity Resolution 20

Combining Global and Local Merges in Logic-based Entity Reso...

引用

20th International Conference on Principles of Knowledge Representation and Reasoning, KR 2023

作者： Bienvenu, Meghyn Cima, Gianluca Gutiérrez-Basulto, Víctor Ibáñez-García, Yazmín Univ. Bordeaux CNRS Bordeaux INP LaBRI UMR 5800 France Department of Computer Control and Management Engineering Sapienza University of Rome Italy School of Computer Science and Informatics Cardiff University United Kingdom

ISBN: (纸本)9781956792027

In the recently proposed LACE framework for collective entity resolution, logical rules and constraints are used to identify pairs of entity references (e.g. author or paper ids) that denote the same entity. This identification is global: all occurrences of those entity references (possibly across multiple database tuples) are deemed equal and can be merged. By contrast, a local form of merge is often more natural when identifying pairs of data values, e.g. some occurrences of 'J. Smith' may be equated with 'Joe Smith', while others should merge with 'Jane Smith'. This motivates us to extend LACE with local merges of values and explore the computational properties of the resulting formalism. © 2023 Proceedings of the International Conference on Knowledge Representation and Reasoning. All rights reserved

关键词： Knowledge representation

来源：评论

学校读者我要写书评

暂无评论

A Machine Learning Approach for Expression Detection in Healthcare Monitoring systems

引用

computers, Materials & Continua 2021年第5期67卷 2123-2139页

作者： Muhammad Kashif Ayyaz Hussain Asim Munir Abdul Basit Siddiqui AaqifAfzaal Abbasi Muhammad Aakif Arif Jamal Malik Fayez Eid Alazemi Oh-Young Song Department of Computer Science&Software Engineering International Islamic UniversityIslamabad44000Pakistan Department of Computer Science Quaid-i-Azam UniversityIslamabad44000Pakistan Department of Computer Science Capital University of Science&TechnologyIslamabad44000Pakistan Department of Software Engineering Foundation UniversityIslamabad44000Pakistan Department of Computer Science Abdul Wali Khan UniversityMardan23200Pakistan Department of Computer Science and Information Systems College of Business StudiesPAAET12062Kuwait Department of Software Sejong UniversitySeoul05006Korea

Expression detection plays a vital role to determine the patient’s condition in healthcare *** helps the monitoring teams to respond swiftly in case of *** to the lack of suitable methods,results are often compromised in an unconstrained environment because of pose,scale,occlusion and illumination variations in the image of the face of the patient.A novel patch-based multiple local binary patterns(LBP)feature extraction technique is proposed for analyzing human behavior using facial expression *** consists of three-patch[TPLBP]and four-patch LBPs[FPLBP]based feature engineering *** representation is encoded from local patch statistics using these *** and FPLBP capture information that is encoded to find likenesses between adjacent patches of pixels by using short bit strings contrary to pixel-based *** images are transformed into the frequency domain using a discrete cosine transform(DCT).Most discriminant features extracted from coded DCT images are combined to generate a feature *** vector machine(SVM),k-nearest neighbor(KNN),and Naïve Bayes(NB)are used for the classification of facial expressions using selected *** experimentation is performed to analyze human behavior by considering standard extended Cohn Kanade(CK+)and Oulu–CASIA *** demonstrate that the proposed methodology outperforms the other techniques used for comparison.

关键词： Detection expressions gestures analytics pain patch-based local binary descriptor discrete cosine transform healthcare

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Models for Brain Signal Classification: A Focus on EEG Analysis in Epilepsy Cases 2

Machine Learning Models for Brain Signal Classification: A F...

引用

2nd International Conference on Cyber Resilience, ICCR 2024

作者： Ali, Ali M Nashwan, Shadi Al-Qerem, Ahmad Almomani, Ammar Sakhnini, Mahmoud Al Aldweesh, Amjad AlAhliyya Amman University Communications And Computer Engineering Department Faculty Of Engineering Amman19328 Jordan Cybersecurity Department Faculty Of Information Technology Amman11831 Jordan Zarqa University Computer Science Department Faculty Of Information Technology Zarqa13110 Jordan School Of Computing Skyline University College Sharjah United Arab Emirates AlHuson University College AlBalqa Applied University I.T. Department Irbid Jordan Shaqra University College Of Computing And It Shaqra Saudi Arabia

ISBN: (纸本)9798350394962

Electroencephalography (EEG) provides critical insights into brain function and neurological disorders. However, analyzing and classifying EEG signals remains challenging. Manual review is time-consuming, prone to bias, and lacks scalability. Hence, developing automated EEG classification has become vital for clinical assessment and treatment of epilepsy. This study presents a machine learning framework for classifying normal and epileptic EEG recordings. The approach involves feature extraction using discrete wavelet transform (DWT) and statistical methods, feature selection to identify the most discriminative attributes, followed by classification across multiple algorithms. Specifically, DWT decomposes non-stationary signals for time-frequency representation. Principal component analysis and cosine similarity assist in selecting robust features. Supervised classifiers including naive Bayes, decision trees, neural networks, k-nearest neighbors, random forest, and support vector machines categorize the signals. Results demonstrate 100% accuracy with neural networks, indicating highly reliable automated classification is achievable. By comparing multiple techniques, the optimal machine learning pipeline emerges. This epilepsy EEG classification framework demonstrates the potential for AI to significantly improve screening, diagnosis, treatment, and management of neurological disorders. Ongoing research aims to further enhance efficiency, scalability, and real-time capabilities. © 2024 IEEE.

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Toward a Secure Industrial Wireless Body Area Network Focusing MAC Layer Protocols: An Analytical Review

引用

IEEE Transactions on Industrial Informatics 2023年第2期19卷 2028-2038页

作者： Javadpour, Amir Sangaiah, Arun Kumar Ja'Fari, Forough Pinto, Pedro Memarzadeh-Tehran, Hamidreza Rezaei, Samira Saghafi, Fatemeh Harbin Institute of Technology Department of Computer Science and Technology Shenzhen150001 China Instituto Politécnico de Viana Do Castelo ADiT-Lab Electrotechnics and Telecommunications Department Porto4900-347 Portugal National Yunlin University of Science and Technology International Graduate Institute of Ai Yunlin64002 Taiwan Sharif University of Technology Department of Computer Engineering Tehran11365-8639 Iran University of Tehran Faculty of New Sciences and Technologies Tehran14155-6619 Iran University of Groningen Bernoulli Institute for Mathematics and Computer Science Groningen9712 CP Netherlands University of Tehran Faculty of Management Tehran14155-6619 Iran

Monitoring security and quality of service is essential, due to the rapid growth of the number of nodes in wireless networks. In healthcare/industrial environments, especially in wireless body area networks (WBANs), this is even more important. Because the delays and errors can directly affect patients'/scientists' health. To increase the Monitoring Quality of Services (MQoS) in WBANs, a secure medium access control (MAC) protocol needs to be developed to provide optimal services. This article provides a comprehensive review of MAC protocols in WBANs with a technical security analysis approach. Time-based, contention-based, and hybrid protocols are compared in this article, regarding MQoS and their security vulnerabilities. We have considered delay, packet loss, and energy consumption as performance evaluation criteria in WBANs, which may be degraded under a cyberattack. This work shows that there is a research gap in the literature, which is the failure of covering security and privacy issues in the MAC layer protocols. © 2005-2012 IEEE.

关键词： Time division multiple access

来源：评论

学校读者我要写书评

暂无评论

Energy-Optimized Routing Protocol for Heterogeneous Wireless Body Area Network Using Optimization Algorithm 4th

Energy-Optimized Routing Protocol for Heterogeneous Wireless...

引用

4th Doctoral Symposium on Computational Intelligence, DoSCI 2023

作者： Bedi, Pradeep Das, Sanjoy Goyal, S.B. Department of Computer Science Regional Campus Manipur Indira Gandhi National Tribal University Madhya Pradesh Amarkantak India Department of Computer Science and Engineering Galgotias University Uttar Pradesh Greater Noida India Faculty of Information Technology City University Petaling Jaya46100 Malaysia

ISBN: (纸本)9789819937158

A WBAN consists of a network of multipurpose nodes with limited storage and limited lifetimes. Keeping these nodes running on their limited battery power is a difficult undertaking. As a result, in any WSN-IoT network, energy saving is of major concern. Due to the ever-changing conditions of the actual world, WBAN must be designed to accommodate a diverse set of nodes if it is to survive for any length of time. The main difficulty, out of all those that are possible, is energy consumption during data communication among these scarce resource sensors. As the signal is continuously sensed by these sensors, messages are sent to other nodes. Therefore, it is necessary to maximize the use of energy in WBAN. This paper’s major goal is to maximize the effectiveness of optimization techniques. With the advantages of double Q-learning for cluster head selection and cluster updation, a cluster-based routing protocol is designed and presented for WBAN. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.

关键词： Routing protocols

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：