检索结果-内蒙古大学图书馆

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

作者： Chen, Yuan Zhong, Chongju Huang, Pinyi Cai, Wangyang Wang, Lei School of Computer Science and Engineering Central South University China School of Computer and Communication Engineering Changsha University of Science and Technology China

ISBN: (纸本)9798350368741

Micro-expression (ME) recognition holds great potential for revealing true human emotions. A significant barrier to effective ME recognition is the lack of sufficient annotated ME video data because MEs are subtle and involuntary facial expressions that are very hard to capture. To address this issue, data augmentation techniques, such as ME migration based on a driven video, have been employed to enrich training samples. Considering that MEs can be complex facial movements involving multiple action unit (AU) changes, we propose a novel ME generation approach that enables the creation of more realistic facial sequences by fusing MEs from multiple videos rather than just single driven video. To enhance the effectiveness of multi-sequence ME transfer, we adapt the thin plate spline motion model and improve traditional face alignment methods to better suit the model, facilitating multi-sequence driven ME generation. In our experiments, we conduct a downstream ME recognition task using models trained on our augmented ME sequences to demonstrate the effectiveness of our approach on the SAMM, SMIC, and CASME II datasets. The results confirm that our proposed approach outperforms state-of-the-art (SOTA) augmentation and generation methods in terms of F1 score and recognition accuracy. © 2025 IEEE.

关键词： face alignment micro-expression generation thin plate spline motion model video fusion

来源：评论

学校读者我要写书评

暂无评论

A Multi-Access Edge Computing Approach to Intelligent Tutoring Systems for Real-Time Adaptive Learning

引用

International Journal of Information technology (Singapore) 2025年第4期17.0卷 2117-2128页

作者： Singh, Ramesh Konyak, Chenlep Yakha Longkumer, Akangjungshi Computer Science and Engineering School of Engineering and Technology Nagaland University Meriema Nagaland Kohima 797004 India

Intelligent tutorial systems (ITS) lack real-time adaptive learning features to provide better learning experiences to the learners. To address this problem, a novel flexible learning system (FLS) is proposed which supports real-time adaptive learning in ITS using a multi-access edge computing (MEC) approach. The adaptive features create personalized learning styles and content tailored to each student’s context, facilitating one-to-one learning. The MEC environment allows the FLS to process tasks for content generation with real-time responses while optimizing network bandwidth usage. The proposed FLS reduced task processing time by 94%, enabling real-time responses that facilitate flexible learning experiences. © Bharati Vidyapeeth's Institute of computer Applications and Management 2025.

关键词： E-learning Intelligent tutor system Learning style Multi-access edge computing

来源：评论

学校读者我要写书评

暂无评论

Optimized Neural Network for Automated Cardiovascular Disease Classification: Meta-heuristic Enabled Model

引用

SN computer Science 2025年第1期6卷 1-14页

作者： Kusuma, S. Jothi, K.R. Department of Computer Science & Engineering (Data Science) Madanapalle Institute of Technology & Science Madanapalle India School of Computer Science and Engineering Vellore Institute of Technology Vellore India

Cardiovascular disease remains a major issue for mortality and morbidity, making accurate classification crucial. This paper introduces a novel heart disease classification model utilizing Electrocardiogram (ECG) signals. The model consists of two main steps: feature extraction using a refined Lion Algorithm with New Mutated evaluation (LA-NM) and classification through an ensemble approach combined with a Neural Network (NN). The proposed model significantly outperforms traditional classifiers. At 80% training data, it achieves accuracy improvements of 8.4% to 3.11% over Neural Network (NN), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and ensemble models. At 90% training data, it shows precision improvements ranging from 1.03% to 0.31% better than the comparison models and specificity improvements ranging from 13.92% to 0.52%. These results underscore the effectiveness of the model and suggest its potential for further development in personalized treatment recommendations based on patient history. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Ensemble models Feature extraction Heart disease classification Neural network Optimization

来源：评论

学校读者我要写书评

暂无评论

Proposed AES-based Obfuscation and Encryption Model for Cloud Data Security 1

Proposed AES-based Obfuscation and Encryption Model for Clou...

引用

1st International Conference on Intelligent Systems and Computational Networks, ICISCN 2025

作者： Singh, Prabhdeep Singh, Pawan Agarwal, Abhay Kumar Kumar, Anil Amity School of Engineering and Technology Department of Computer Science and Engineering Lucknow India Kamla Nehru Institute of Technology Department of Computer Science and Engineering Sultanpur India Amity University Amity School of Engineering and Technology Lucknow India

ISBN: (纸本)9798331529246

Cloud Computing (CC) offers a diverse range of services along with huge data storage across a network. CC has collaborated with varied emerging technologies like IoT because of its numerous advantages. Despite CC's benefits, cloud-based servers are not always trustworthy in real-world scenarios. Users' privacy will be compromised if they send their data straight to cloud platforms since crucial information will be exposed to cloud servers. Thereby, this work develops a data security model in the cloud with two stages, proposed secured data storage and proxy re-encryption. Initially, the data is stored securely in the cloud based on improved AES-based encryption as well as obfuscation-based encryption. Based on the request from the data user, proxy re-encryption is done using Kronecker product-based encryption. The enhanced AES achieved the lowest value of 0.025 for the CCA attack with a key size of 128 bits. Thus, the final encoded key is obtained. © 2025 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Federated ADMM

IEEE Networking Letters

引用

IEEE Networking Letters 2025年第1期7卷 11-15页

作者： Azimi-Abarghouyi, Seyed Mohammad Bastianello, Nicola Johansson, Karl H. Fodor, Viktoria KTH Royal Institute of Technology School of Electrical Engineering and Computer Science and Digital Futures Stockholm10044 Sweden

In this letter, we depart from the widely-used gradient descent-based hierarchical federated learning (FL) algorithms to develop a novel hierarchical FL framework based on the alternating direction method of multipliers (ADMM), leveraging a network architecture consisting of a single cloud server and multiple edge servers, where each edge server is dedicated to a specific client set. Within this framework, we propose two novel FL algorithms, which both use ADMM in the top layer: one that employs ADMM in the lower layer and another that uses the conventional gradient descent-based approach. The proposed framework enhances privacy, and experiments demonstrate the superiority of the proposed algorithms compared to the conventional algorithms in terms of learning convergence and accuracy. Additionally, gradient descent on the lower layer performs well even if the number of local steps is very limited, while ADMM on both layers lead to better performance otherwise. © 2019 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Towards Maximizing Semantic Coverage for Image-Text Retrieval

Towards Maximizing Semantic Coverage for Image-Text Retrieva...

引用

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

作者： Xu, Junhao Liu, Zheng Chen, Zhumin Shen, Fei School of Computer Science and Technology Shandong University China School of Computer Science and Technology Shandong University of Finance and Economics China School of Computer Science and Engineering Nanjing University of Science and Technology China

ISBN: (纸本)9798350368741

To establish semantic associations between images and texts, existing Image-Text Retrieval (ITR) methods primarily focus on fixed-scale fragments, which only identify explicit semantic categories. Consequently, semantic coverage is constrained, leading to the omission of certain semantic associations. To enlarge the semantic coverage, we propose the Semantic Coverage-Aware Network (SCA-Net). First, explicit semantic categories are identified by SCA-Net through analyzing the semantic membership of visual and textual fragments, thereby establishing more precise explicit semantic associations. Second, implicit semantic categories are identified by SCA-Net via adaptively aggregating visual and textual fragments across various scales using a co-occurrence-aware router, thereby significantly expanding the semantic coverage and establishing complete semantic associations. Third, image-text similarity is calculated using the attention mechanism over a broader range of semantic coverage. Extensive experiments demonstrate that SCA-Net significantly enhances ITR performance compared to state-of-the-art methods by maximizing semantic coverage. © 2025 IEEE.

关键词： Co-occurrence-aware router Image-text retrieval Semantic association Semantic category Semantic coverage

来源：评论

学校读者我要写书评

暂无评论

Skeleton-Based Action Recognition Using Graph Convolutional Network with Pose Correction and Channel Topology Refinement

引用

computers, Materials & Continua 2025年第4期83卷 701-718页

作者： Yuxin Gao Xiaodong Duan Qiguo Dai School of Engineering Guangzhou College of Technology and BusinessFoshan528138China School of Computer Science and Engineering Dalian Minzu UniversityDalian116000China SEAC key Laboratory of Big Data Applied Technology Dalian Minzu UniversityDalian116000China

Graph convolutional network(GCN)as an essential tool in human action recognition tasks have achieved excellent performance in previous ***,most current skeleton-based action recognition using GCN methods use a shared topology,which cannot flexibly adapt to the diverse correlations between joints under different motion *** video-shooting angle or the occlusion of the body parts may bring about errors when extracting the human pose coordinates with estimation *** this work,we propose a novel graph convolutional learning framework,called PCCTR-GCN,which integrates pose correction and channel topology refinement for skeleton-based human action ***,a pose correction module(PCM)is introduced,which corrects the pose coordinates of the input network to reduce the error in pose feature ***,channel topology refinement graph convolution(CTR-GC)is employed,which can dynamically learn the topology features and aggregate joint features in different channel dimensions so as to enhance the performance of graph convolution networks in feature ***,considering that the joint stream and bone stream of skeleton data and their dynamic information are also important for distinguishing different actions,we employ a multi-stream data fusion approach to improve the network’s recognition *** evaluate the model using top-1 and top-5 classification *** the benchmark datasets iMiGUE and Kinetics,the top-1 classification accuracy reaches 55.08%and 36.5%,respectively,while the top-5 classification accuracy reaches 89.98%and 59.2%,*** the NTU dataset,for the two benchmark RGB+Dsettings(X-Sub and X-View),the classification accuracy achieves 89.7%and 95.4%,respectively.

关键词： Pose correction multi-stream fusion GCN action recognition

来源：评论

学校读者我要写书评

暂无评论

PAH2T-Former: Paired-Attention Hybrid Hierarchical Transformer for Synergistically Enhanced FMT Reconstruction Quality and Efficiency

IEEE Transactions on Computational Imaging

引用

IEEE Transactions on Computational Imaging 2025年 11卷 536-545页

作者： Zhang, Peng Liu, Xingyu Xue, Qianqian Shang, Yu Liu, Chen Chen, Ruhao Gao, Honglei Liang, Jiye Wang, Wenjian Zhang, Guanglei Shanxi University School of Computer and Information Technology Taiyuan030006 China Dongguan University of Technology School of Life and Health Technology Dongguan523808 China Beihang University School of Biological Science and Medical Engineering Beijing100191 China

Fluorescence molecular tomography (FMT) is a sensitive optical imaging technique that can achieve three-dimensional (3D) tomographic images at the molecular and cellular levels. However, reconstructing the internal 3D distribution of fluorescent targets from surface two-dimensional (2D) fluorescence projection data remains a challenging task. In recent years, deep learning-based FMT reconstruction has received considerable attention, demonstrating superior performance compared to conventional methods, particularly combined with Transformers. Unlike convolutional architectures that emphasize local context, Transformers leverage self-attention mechanisms to excel at capturing long-range dependencies, thereby enhancing FMT reconstruction accuracy. Nevertheless, the quadratic computational complexity of self-attention poses a bottleneck, particularly pertinent in 3D FMT reconstructions. This paper aims to propose a novel Transformer-based FMT reconstruction algorithm that not only delivers high-quality reconstruction accuracy but also maintains excellent performance in efficiency and inference speed. The key design involves introducing a novel Spatial-Channel Paired Attention Module (SC-PAM), which employs a pair of interdependent branches based on spatial and channel attention, thus effectively learn discriminative features in both spatial and channel domains, meanwhile exhibiting linear complexity relative to the input projection size. Furthermore, to facilitate data transmission between the spatial and channel branches, we share the weights of the query and key mapping functions, which provides a complementary paired attention without elevating complexity. Extensive evaluations through numerical simulations and in vivo experiments were performed to validate effectiveness of the proposed model. The results show that our PAH2T-Former method achieves the highest Dice while reducing model parameters and complexity. © 2015 IEEE.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

UAV-Assisted Zero Knowledge Model Proof for Generative AI: A Multiagent Deep Reinforcement Learning Approach

引用

IEEE Internet of Things Journal 2025年第10期12卷 13441-13454页

作者： Hao, Min Shang, Chen Wang, Siming Jiang, Wenchao Nie, Jiangtian Information Systems Technology and Design Singapore University of Technology and Design 487372 Singapore School of Computer Science and Engineering Nanyang Technological University 638788 Singapore

As more users seek generative AI (GAI) models to enhance work efficiency, GAI and Model-as-a-Service will drive transformative changes and upgrades across all industries. However, when users utilize GAI models provided by the service provider, they cannot be certain that the model’s quality matches the provider’s claims. Considering the need to protect intellectual property, the service provider will not disclose model details for user verification. To this end, we take the Internet of Vehicles as research background, proposing a zero knowledge model proof architecture based on UAVs. We also introduce a multiagent reinforcement learning algorithm to optimize the verification process. In specific, we first propose a verification scheme for the key operations of generative adversarial networks based on noninteractive zero knowledge proof. The zero knowledge proof architecture ensures that model parameters cannot be stolen during the verification process. After that, we propose an Age of Verification (AoV) metric to ensure the timeliness and freshness of zero knowledge proof. We also construct a tradeoff optimization problem between the energy consumption of UAV as a verifier and the AoV of edge servers as service providers, and transform the problem based on Lyapunov optimization theory. Following that, we propose an enhanced multiagent proximal policy optimization algorithm to enable the collaborative verification of edge servers by multiple UAVs. The algorithm simulation results demonstrate that the reward value of our proposed algorithm is over 10% higher than that of the standard algorithm, with a faster and more stable overall convergence speed. Additionally, the zero knowledge proof performance test results indicate that the verification delay in our proposed architecture is less than 500 ms during the verification phase, meeting practical requirements. © 2014 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Crack-DETR: Complex Pavement Crack Detection by Multifrequency Feature Extraction and Fusion

引用

IEEE Sensors Journal 2025年第9期25卷 16349-16360页

作者： Liu, Guoxi Liu, Guozhi Chai, Xinzhuo Li, Lecheng Dai, Fei Huang, Bi Southwest Forestry University School of Big Data and Intelligent Engineering Kunming650233 China South China University of Technology School of Computer Science and Engineering Guangdong Guangzhou510006 China

Automated detection of pavement cracks plays a crucial role in road maintenance and traffic safety. However, pavement crack detection under noisy conditions is challenging due to the complex expression forms of pavement cracks and the presence of multitype background noise. In this article, we propose Crack-DETR, an efficient pavement crack detection deep learning network, that can detect complex pavement cracks from complex background noise. In this network, we first propose a multifrequency feature extraction and fusion (MFEF) block to enhance pavement crack features and suppress background noise from both high-frequency and low-frequency perspectives in two stages: multifrequency feature extraction and multifrequency feature fusion. In the multifrequency feature extraction stage, we design a high-frequency attention mechanism to extract high-frequency features from high-level features. Then, we propose a low-frequency attention mechanism to capture low-frequency features of high-level features. In the multifrequency feature fusion stage, we fuse the high- and low-frequency features through intrafrequency feature fusion and cross-frequency feature fusion. Second, we propose an evaluating metric, diagonal intersection over union (DIoU), to enhance the loss function of Crack-DETR, improving the learning effectiveness of the model. Finally, we conduct extensive experiments on three real-world datasets with different scenarios, and the experimental results show that our method can improve the frames per second (FPS) by about 33.4% and mAP@50 by about 6.78% on average compared with baseline model in the YN dataset. Consequently, Crack-DETR achieves higher accuracy while reducing computational complexity, paving the way for the development of sensor-based real-time pavement crack detection. © 2001-2012 IEEE.

关键词： Feature extraction

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：