检索结果-内蒙古大学图书馆

UPB Scientific Bulletin, Series C: Electrical Engineering and computer science 2024年第1期86卷 155-166页

作者： Alwan, Hiba Basim Ku-Mahamud, Ku Ruhana Dept. of Computer Science University of Technology Iraq School of Computing Universiti Utara Malaysia Malaysia

Biometrics systems utilizing hand geometry, fingerprint, iris, face, palm print, voice, gesture, and palm print have been utilised for authentication purposes. Through these templates, the face template is suggested as the strongest. However, problems still exist in face-matching implementation which affects biometric security. Thus, in this paper, a new cancellable face image algorithm using the Convolution Neural Network (CNN) feature extraction technique and Winner-Takes-All hashing method is proposed. The algorithm is to overcome the problem of matching which will enhance its security. ORL and Yale datasets were utilized in the evaluation of the proposed algorithm. Several common algorithms were utilised in the comparison based on Equal Error Ratio (ERR) measurement. The outcomes displayed the verification performance of the unimportant differences with EER = 0.95% and 3.25% for the extended version of the face dataset for the Yale and ORL face datasets, respectively. The acceptable implementation of the proposed algorithm demonstrates that it can be used for the security purpose of biometric trait tasks. © 2024, Politechnica University of Bucharest. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Seeking a Sense of Meaning and Companionship in Life: Informal Learning on Douyin Among Chinese Older Adults

引用

Proceedings of the ACM on Human-computer Interaction 2025年第2期9卷 1-24页

作者： Qian, Zhigu Fu, Jiaojiao Zhou, Yangfan School of Computer Science Fudan University Shanghai Key Laboratory of Intelligent Information Processing China School of Information Science and Engineering East China University of Science and Technology China

This study examines how Chinese older adults leverage Douyin, a short video platform, for informal learning purposes, analyzing their usage patterns, motivations, and encountered challenges. Although Douyin was not explicitly designed with educational features, it has emerged as a significant informal learning platform for this demographic. Through a qualitative investigation comprising participant observations and semi-structured interviews with 17 participants, we reveal the distinctive learning experience that Douyin facilitates. The platform’s unique combination of short-form videos, live streaming capabilities, and interactive community features creates an engaging learning environment that particularly resonates with older adults. Our findings demonstrate that participants derive substantial social-emotional benefits beyond knowledge acquisition, including strengthened social connections, enhanced companionship, and a reinforced sense of purpose through their Douyin engagement. These social-emotional aspects emerge as crucial factors driving older adults’ selection of Douyin as their preferred learning platform. By providing comprehensive insights into older adults’ engagement with digital platforms for lifelong learning, this research offers valuable implications for HCI, particularly in understanding how technology can be optimized to support the learning needs of older populations. © 2025 Association for Computing Machinery. All rights reserved.

关键词： Social aspects

来源：评论

学校读者我要写书评

暂无评论

Clustering Algorithm of Remote Sensing Image Change Detection Based on Improved U-Net Network 3

Clustering Algorithm of Remote Sensing Image Change Detectio...

引用

3rd International Conference on Data Analytics, Computing and Artificial Intelligence, ICDACAI 2024

作者： Peng, Song School of Computer Science and Technology University of Science and Technology of China Anhui Province Hefei China

ISBN: (纸本)9798331519032

Disaster assessment, urban planning and environmental monitoring urgently need to be combined with remote sensing image change detection to play a role. An improved U-Net network combined with clustering algorithm is proposed for remote sensing image change detection. In this paper, firstly, the U-Net network is improved to enhance its feature extraction ability. Next, this paper takes the improved U-Net output feature map as the input of clustering algorithm, and uses K-means clustering algorithm, density-based noisy application spatial clustering (DBSCAN) and Mean Shift clustering to process the change detection results. The final experimental results show that the improved U-Net network combined with K-means algorithm has achieved remarkable performance improvement in remote sensing image change detection, with the maximum OA of 98%, the maximum Kappa coefficient of 0.94 and the maximum BAI of 0.98. © 2024 IEEE.

关键词： Change detection

来源：评论

学校读者我要写书评

暂无评论

Enhanced Privacy Preserving for Social Networks Relational Data Based on Personalized Differential Privacy

引用

Chinese Journal of Electronics 2022年第4期31卷 741-751页

作者： KANG Haiyan JI Yuanrui ZHANG Shuxuan School of Information Management Beijing Information Science and Technology University Computer School Beijing Information Science and Technology University

With the popularization and development of social software, more and more people join the social network, which produces a lot of valuable information, but also contains plenty of sensitive privacy information. To achieve the personalized privacy protection of massive social network relational data, a privacy enhancement method for social networks relational data based on personalized differential privacy is proposed. And a dimensionality reduction segmentation sampling(DRS-S)algorithm is proposed to implement this method. First, in order to solve the problem of inefficiency caused by the excessive amount of data in social networks, dimension reduction and segmentation are carried out to divide the data into groups. According to the privacy protection requirements of different users, we adopt sampling method to protect users with different privacy requirements at different levels, so as to realize personalized different privacy. After that, the noise is added to the protected data to satisfy the privacy budget. Then publish the social network data. Finally, the proposed algorithm is compared with the traditional personalized differential privacy(PDP) algorithm and privacy preserving approach based on clustering and noise(PBCN) in real data set, the experimental results demonstrate that the quality of privacy protection and data availability of DRS-S are better than that of PDP algorithm and PBCN algorithm.

关键词： Social network Privacy preserving Dimensionality reduction segmentation Personalized differential privacy DRS-S algorithm

来源：评论

学校读者我要写书评

暂无评论

Specialized Accelerator for Arbitrary-Dimension Matrix Operations in Deep Learning 17

Specialized Accelerator for Arbitrary-Dimension Matrix Opera...

引用

17th IEEE International Conference on computer Research and Development, ICCRD 2025

作者： Liu, Fei Liu, Zhouquan Zhang, Jing Lai, Mingche Guo, Hui Huang, Libo National University of Defense Technology College of Computer Science and Technology Changsha China

ISBN: (纸本)9798331531881

This paper addresses the significant challenge of executing inference tasks involving General Matrix Multiplication (GEMM) in deep neural networks(DNN) on resource-constrained edge systems. While previous research has introduced various accelerator architectures for GEMM, these solutions primarily support matrix multiplications of varying dimensions at the software level rather than the hardware level. We propose an optimization of data reuse and memory access patterns through the integration of the outer product method. Building on this, we developed an optimized workload allocation strategy to further minimize data transfer overhead and balance computational resource utilization. Additionally, a heterogeneous hardware module named the Matrix Acceleration Engine (MAE) has been designed and integrated into the RISC-V processor to accelerate prevalent GEMM operations in deep neural networks. Compared to traditional instruction sets, the RV-MAE processor achieves a speedup of up to 194.74× in matrix operations and up to 20.29× in convolutional neural network tasks. Compared to advanced matrix accelerators, the performance is improved by 5.33×. Furthermore, the MAE, evaluated in a 28nm CMOS process, exhibits an area of 9742.18μm2and a power consumption of 2.9mW, thus meeting the low-power demands of edge computing. © 2025 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

A proposed deep learning model for multichannel ECG noise reduction

引用

Discover Artificial Intelligence 2025年第1期5卷 1-16页

作者： Maurya, Jay Prakash Manoria, Manish Joshi, Sunil Department of Computer Science Engineering Samrat Ashok Technological Institute Madhya Pradesh Vidisha India Director & Professor Department of Computer Science Engineering Rungta College of Engineering & Technology Bhilai Chhattisgarh Durg India

Heart disease is a critical concern of healthcare for everyone in today’s era. An effective and noninvasive indication of heart disease is an electrocardiogram (ECG). Understanding regular ECG signal patterns and comparisons with irregular, patterns may help to identify the serious nature of heart diseases such as arrhythmia. Comparison of ECG signal patterns is very difficult manually, and machine-based interpretation is a demand of society. Errors in ECG interpretation might result from noise contamination of ECG signals. ECG pretreatment from noise is therefore needed for precise analysis. This article proposed a novel deep learning-based solution for multichannel ECG noise reduction, through utilizing the capabilities of fully convolutional neural network along with the Jacobin regularization to ensure confining and preserving local information. Cascaded layered approach was framed in encoder and decoder sections of model where denoising and reconstruction process worked and compared on standard performance parameters with recent denoising deep autoencoders. This proposed work FCN-DAE with Jacobin regularization uses the noise stress test database (NSTDB) for the noise signal of ECG data sourced from the PhysioNet repository. The proposed model achieves 4.763 * 10–2 mv2 for signal space diversity (SSD), 0.288 mv for the median absolute deviation and 1.859 for the root mean square error (RMSE) in the conducted experiment. The experimental findings demonstrate that complex noise from the ECG signal may be removed up to 97.02%. © The Author(s) 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

iLIAC: An approach of identifying dissimilar groups on unstructured numerical image dataset using improved agglomerative clustering technique

引用

Multimedia Tools and Applications 2024年第39期83卷 86359-86381页

作者： S, Sreedhar Kumar Ahmed, Syed Thouheed Fathima, Afifa Salsabil Mathivanan, Sandeep Kumar Jayagopal, Prabhu Saif, Abdu Gupta, Sachin Kumar Sinha, Garima Department of Computer Science and Engineering School of Engineering and Technology CMR University Karnataka Bengaluru India Department of Electrical Engineering Indian Institute of Technology Hyderabad Telangana Kandi India School of Computer Science and Engineering REVA University Bangalore India School of Computer Science and Engineering Galgotias University Uttar Pradesh Greater Noida203201 India School of Computer Science Engineering and Information Systems Vellore Institute of Technology Tamil Nadu Vellore632014 India Department of Communication and Computer Engineering Faculty of Engineering and IT Taiz University Taiz Yemen Katra India Department of Computer Science Engineering Jain University Bangalore India

Unstructured Numerical Image Dataset Separation (UNIDS) method employing an enhanced unsupervised clustering technique. The objective is to delineate an optimal number of distinct groups within the input grayscale (G-S) image content, utilizing the improved limited iteration agglomerative clustering (iLIAC) clustering technique for the separation and enhancement of picture elements. The UNIDS method is structured into two primary stages: partitioning and validation. In the partitioning stage, the UNIDS method identifies an appropriate number of discrete clusters within the grayscale image using the iLIAC technique, eliminating the need for predetermined procedures. Subsequently, the method evaluates the similarity and deviation among data elements within the same group in the resultant image dataset. Additionally, it assesses the proximity and inters severance among clusters in the outcome of the image dataset through the partitioning process. Empirical results indicate that the UNIDS system excels in the spontaneous identification of an optimal number of discrete clusters within the input G-S image. The system demonstrates superior thickness, reduced deviation among data elements within the same cluster, increased inter-separation, and diminished inter-closeness between cluster elements. Furthermore, empirical analysis establishes the superior performance of the UNIDS approach compared to existing clustering techniques. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Anchorless Target Tracking Based on Multilayer Feature Fusion 6

Anchorless Target Tracking Based on Multilayer Feature Fusio...

引用

6th International Conference on Artificial Intelligence and Advanced Manufacturing, AIAM 2024

作者： Peng, Shan Chen, Yungui Zhang, Guangtai Zhou, Zhihang Yang, Lei Department of Computer Science Guangdong University of Science and Technology Dongguan China

ISBN: (纸本)9781837242672

Target tracking is an important part of the field of computer vision, widely used in systems such as unmanned aerial vehicles, automatic driving, sports events, etc., and is closely related to target detection, image segmentation and so on. Aiming at the traditional twin network-based target tracking algorithm which is often prone to tracking failure when facing complex scenes such as object's deformation, scale change, occlusion, etc., this paper proposes a multilayer feature fusion method for fusing deep features and shallow feature information, uniting the feature maps of the multilayer network, composing a continuous feature space, and fusing the high-latitude features of the deeper network under the premise of preserving the details of image features, which enriches the image feature information extracted by the network model, improves the diversity of feature extraction, and also reduces the number of parameters for network training. Experimental results on the VOT dataset show that this method is higher than SiamFC, SiamVGG, DSiam and other methods in tracking accuracy. © The Institution of Engineering & technology 2024.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Network Density Based Centrality for Influential Node Detection in Complex Networks

An Enhanced Network Density Based Centrality for Influential...

引用

2024 IEEE Region 10 Conference, TENCON 2024

作者： Mitra, Sourav Kumar Chakraborty, Susanta Kule, Malay Indian Institute of Engineering Science and Technology Department of Computer Science and Technology Shibpur Howrah India

ISBN: (纸本)9798350350821

Comprehending the dynamics of disease or information diffusion requires influential node detection in complex networks. Numerous centrality measures were developed for calculating the spreading capacity of a node. However, they fail to produce a high monotonic ranking of spreaders, which is a highly desired property. Two new measures, local density and extended local density are introduced in this paper to calculate the spreading impact of every node using the density of edges in the neighborhood of the node and the node degree. Our experimental results demonstrate that both of our proposed metrics offer a more consistent monotonic result for node ranking and better rank distribution than other centrality measures. © 2024 IEEE.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

PAGE: Parametric Generative Explainer for Graph Neural Network 27

PAGE: Parametric Generative Explainer for Graph Neural Netwo...

引用

27th European Conference on Artificial Intelligence, ECAI 2024

作者： Qiu, Yang Liu, Wei Wang, Jun Li, Ruixuan School of Computer Science and Technology Huazhong University of Science and Technology China iWudao Tech China

ISBN: (纸本)9781643685489

This article introduces PAGE, a parameterized generative interpretive framework. PAGE is capable of providing faithful explanations for any graph neural network without necessitating prior knowledge or internal details. Specifically, we train the autoencoder to generate explanatory substructures by designing appropriate training strategy. Due to the dimensionality reduction of features in the latent space of the autoencoder, it becomes easier to extract causal features leading to the model's output, which can be easily employed to generate explanations. To accomplish this, we introduce an additional discriminator to capture the causality between latent causal features and the model's output. By designing appropriate optimization objectives, the well-trained discriminator can be employed to constrain the encoder in generating enhanced causal features. Finally, these features are mapped to substructures of the input graph through the decoder to serve as explanations. Compared to existing methods, PAGE operates at the sample scale rather than nodes or edges, eliminating the need for perturbation or encoding processes as seen in previous methods. Experimental results on both artificially synthesized and real-world datasets demonstrate that our approach not only exhibits the highest faithfulness and accuracy but also significantly outperforms baseline models in terms of efficiency. © 2024 The Authors.

关键词： Graph neural networks

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

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

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：