检索结果-内蒙古大学图书馆

2024 4th International Conference on Intelligent Traffic Systems and Smart City, ITSSC 2024

作者： Li, Wenqiang Sheng, Hu Geng, Na Yang, Xili Xu, Chuanxin School of Computer and Communication Engineering Dalian Jiaotong University Liaoning Province Dalian116028 China

ISBN: (纸本)9781510686304

In response to issues of low detection accuracy and poor robustness caused by missed detections of small objects and occlusions among dense objects in road scenes, a YOLOv8-RC object detection algorithm is proposed. First, by introducing receptive field attention convolution, the network can flexibly adjust the level of attention to targets of different scales, thereby enhancing the feature extraction capability of the backbone network and improving detection accuracy. Secondly, a lightweight CARAFA is used to replace the upsampling operator in the neck network, which reduces the number of parameters while improving detection accuracy. Finally, by adding a small object detection layer, local feature extraction is performed on larger-sized feature maps, thereby improving the accuracy of small object detection. Experimental results show that on a dataset excerpted from BDD100K, the average precision of the YOLOv8-RC algorithm model in this study increased by 5.9% compared to the model before improvement, with the model's parameter count reduced by 0.05M. © 2025 SPIE.

关键词： Object detection Roads Target detection Detection and tracking algorithms Performance modeling Feature extraction Data modeling Small targets Convolution Education and training

来源：评论

学校读者我要写书评

暂无评论

Neighbor-relation aware low-rank multi-view subspace clustering

引用

Multimedia Systems 2025年第3期31卷 1-16页

作者： Jia, Hongjie Wang, Tengteng Song, Heping School of Computer Science and Communication Engineering Jiangsu University Zhenjiang212013 China Jiangsu Engineering Research Center of Big Data Ubiquitous Perception and Intelligent Agriculture Applications Zhenjiang212013 China

Low-rank multi-view subspace clustering seeks to uncover complex relationships in high-dimensional data by leveraging multiple views and low-rank representations. However, two key challenges persist: (1) Existing methods emphasize low-rank consistency through nuclear norm regularization but often overlook the unique low-rank features of individual views. (2) Constructing the affinity matrix directly from the coefficient matrix can introduce excessive connections, degrading clustering performance. To address these challenges, we propose neighbor-relation aware low-rank multi-view subspace clustering (NLMSC), a unified model that integrates view-specific low-rank structure learning with neighbor-aware affinity fusion. NLMSC decomposes the coefficient matrix into smaller factor matrices to capture the underlying low-rank structures of each view and constructs a sparser, more interpretable affinity matrix by reliable neighbor selection. These affinity matrices are then fused into a block-diagonal consensus graph through a self-weighting strategy for spectral clustering. Extensive experiments on diverse multi-view datasets demonstrate NLMSC’s superior performance and effectiveness in challenging clustering tasks. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Low-rank structure Multi-view clustering Neighbor-relation awareness Subspace clustering

来源：评论

学校读者我要写书评

暂无评论

OLMS: A Flexible Online Learning Multi-Path Scheduling Framework

引用

IEEE Transactions on Network Science and engineering 2025年第3期12卷 2277-2291页

作者： Cai, Kechao Chen, Zhuoyue Zhang, Jinbei Lui, John C. S. Shenzhen Campus of Sun Yat-sen University School of Electronics and Communication Engineering Shenzhen510275 China The Chinese University of Hong Kong Department of Computer Science and Engineering Shatin Hong Kong

Over the past decade, there has been a tremendous surge in the inter-connectivity among hosts in networks. Many multi-path transport protocols, such as MPTCP, MPQUIC, and MPRDMA, have emerged to facilitate multi-path data transmissions between pairs of hosts. However, existing packet schedulers in these protocols are quite limited as they neglect the stochastic nature inherent in heterogeneous paths, such as, round-trip time and available bandwidth. Moreover, users have diverse requirements;for instance, some prioritize low latency, while others consistently seek to achieve high bandwidth. In this paper, we propose a flexible Online Learning Multi-path Scheduling (OLMS) framework to schedule packets to multiple paths and meet various user-defined requirements by learning the dynamic characteristics of paths in various applications. Specifically, we consider two types of applications, which are 1) maxRTT constrained and 2) bandwidth constrained, and use OLMS to schedule packets to satisfy the distinct user-defined requirements. Our theoretical analysis demonstrates that OLMS achieves guarantees with sublinear regret and sublinear violation. Furthermore, we implement a prototype of OLMS in MPQUIC and conduct experiments across different scenarios. Our experiments on Mininet show that OLMS enables an 8.42%–18.71% increase in bandwidth utilization in the maxRTT constrained application and negligible violations of user-defined requirements in both applications compared to other schedulers. Additionally, OLMS reduces flow completion times by 4.22%–10.26% compared to other schedulers, all without incurring large overhead. © 2013 IEEE.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

IPT-ILR: Image Pyramid Transformer Coupled With Information Loss Regularization for All-in-One Image Restoration

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年第5期35卷 4341-4356页

作者： Yang, Sai Hu, Bin Liu, Fan Wu, Xiaoxin Ding, Weiping Zhou, Jun Nantong University School of Electrical Engineering and Automation Nantong226019 China Nantong University School of Artificial Intelligence and Computer Science Nantong226019 China Hohai University College of Computer Science and Software Engineering Nanjing210098 China Griffith University School of Information and Communication Technology NathanQLD4111 Australia

All-in-one image restoration has recently developed to be a new research trend in the low-level computer vision field, aiming to tackle multiple image degradation types simultaneously in a unified model. As a typical multi-task learning, existing approaches focus on modeling either the specificity or commonality among different image restoration tasks. To exploit the unique strengths of both worlds, we propose a method of Image Pyramid Transformer coupled with Information Loss Regularization (IPT-ILR), in which the multi-scale architecture structure can excavate more information for multiple restoration tasks concurrently, while the learning strategy can identify the difference among multiple restoration tasks depending on the degree of information loss in each restoration task. Specifically, it first establishes a new Image Pyramid Transformer Network (IPT-Network) to accommodate multiple image restoration tasks. Given original degraded images, the IPT-Network exploits the image pyramid technique to establish a series of images with different scales, which are then restored by transformer-like auto-encoders. Moreover, the restored image on a low-level scale is referenced to assist restoring the degraded image on a high-level scale. Next, Information Loss Regularization (ILR) is presented to optimize the IPT-Network. ILR calculates the average distance between degraded images and their clean counterparts as the weights, which automatically implement different penalties for different image restoration tasks, thus avoiding the short-cut phenomenon for the easy task while encouraging the hard task. Extensive experiments have been conducted with 6 image restoration tasks in the all-in-one setting. The results show our method performs favorably against numerous state-of-the-art methods across most tasks, including image denoising, image deblurring, image dehazing, image deraining, image desnowing, as well as low-light enhancement. © 1991-2012 IEEE.

关键词： Restoration

来源：评论

学校读者我要写书评

暂无评论

AN 8-WAY RF FEED NETWORK FOR 5.3 GHZ C-BAND SAR: DESIGN, SIMULATION, AND ANALYSIS USING STANDARD METHODS

引用

Telecommunications and Radio engineering (English translation of Elektrosvyaz and Radiotekhnika) 2025年第5期84卷 21-43页

作者： Kothapudi, Venkata Kishore Thokala, Niharika Vallabhaneni, Ramyasri Sanaka, Poojitha Kumar, Vijay Pappula, Lakshman Gottapu, Srinivasa Kiran Bhattacharyya, Arka Guntur District Andhra Pradesh Vadlamudi522213 India Department of Communication Engineering School of Electronics Engineering Vellore Institute of Technology Tamil Nadu Vellore632014 India Department of Electrical Electronics and Communication Engineering GITAM School of Technology GITAM University Andhra Pradesh Visakhapatnam530045 India School of Science and Engineering Engineering and Computer Science Department Benedict College 1600 Harden St ColumbiaSC29204 United States

This study proposes the design and analysis of an eight-way power divider for unequal division at 5.3 GHz for C-band frequency. Many transmission line pieces make up the current power divider. These transmission lines were printed on a substrate material (RT/duroid 5880, 1.6 mm thick, permittivity 2.2, and loss tangent 0.0002). The eight-way power divider’s unequal amplitudes are constructed using Taylor’s 1-parameter method to achieve the side projection levels –17 dB down as from the main beam. The proposed power divider consists of using two transmission lines with a 3/4 frequency connected to branched microstriplines, providing excellent isolation and matching between the power divider’s two output ports. This power divider is composed of three stages. A 1:2 power divider is used in the first stage, a 1:4 power divider is used in the second stage, and a 1:8 power divider is used in the third stage. According to the simulation results, 1:2 delivers half of the input power, 1:4 produces one-fourth of the power, and 1:8 delivers one-eighth of the input power. All of these assessments were done with the advanced design system (ADS) simulation program. According to both simulated and measured data, the return loss exceeds –35 dB, and the isolation across the output terminals exceeds –40 dB. The proposed circuit operates effectively at C-band frequencies in terms of return loss and isolation (4 GHz and 8 GHz). © 2025 by Begell House, Inc.

关键词： Integrated circuit design

来源：评论

学校读者我要写书评

暂无评论

GSA-DT: A Malicious Traffic Detection Model Based on Graph Self-Attention Network and Decision Tree

引用

IEEE Transactions on Network and Service Management 2025年第2期22卷 2059-2073页

作者： Cai, Saihua Tang, Han Chen, Jinfu Lv, Tianxiang Zhao, Wenjun Huang, Chunlei Jiangsu University School of Computer Science and Communication Engineering Zhenjiang212013 China Jiangsu Provincial Key Laboratory of Industrial Cyberspace Security Technology Zhenjiang212013 China

Malicious attack has shown a rapid growth in recent years, it is very important to accurately detect malicious traffic to defend against malicious attacks. Compared with machine learning and deep learning technologies, graph convolutional neural network (GCN) achieves better detection results of malicious traffic due to additional consideration of the correlation between network traffic features. However, existing GCN-based detection models suffer from fixed weight assignment, only focusing on local features, lack the ability to model graph structure and relationships as well as having gradient disappearance. To solve these problems, this paper proposes the GSA-DT model based on graph self-attention network and decision tree. GSA-DT first preprocesses the original network traffic to obtain better traffic features and labels, and then uses GCN to extract the topological structure of network traffic as well as capture the correlation relationships among traffic features, where the ReLU activation function is replaced by LeakyReLU to overcome the problems of neuron "death"and gradient disappearance during the training process;It also introduces the self-attention mechanism into GCN to assign larger weights to the key features to reduce the interference of redundant features. Finally, GSA-DT uses decision tree to perform the detection of malicious traffic. Experimental results on five network traffic datasets show that GSA-DT model improves the detection accuracy over 1% on average than seven advanced malicious traffic detection models, and it also performs better in F1-measure, TPR, FPR as well as stability. © 2004-2012 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

A Novel Framework for Detecting Flood Disaster: Spatio-Temporal Fusion and Digital Twin Approach

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 11547-11560页

作者： Lim, Se-Jung Sankaran, K.Sakthidasan Haldorai, Anandakumar Chonnam National University Yeosu Campus School of Electrical and Computer Engineering Jeollanam-do Yeosu-si59626 Korea Republic of Hindustan Institute of Technology and Science Department of Electronics and Communication Engineering Chennai India Sri Eshwar College of Engineering Department of Computer Science and Engineering TamilNadu Coimbatore India

Flood is regarded as common disaster which could cause serious devastation in any country. Typically, it is caused due to precipitation & river runoffs, specifically at the time of excessive rainfall season. Also, by the consequence of global warming and extreme environmental effects, flood becomes a serious issue to the extent on bringing negative impacts to infrastructure and mankind. The technology of sensor network has been used to monitor changes in landcovers and water level fluctuations. Moreover, effective disaster monitoring & notification system in real-time becomes a crucial part which needs to be overcome. For this reason, the proposed methodology is designed aiming at developing natural disaster prediction and monitoring system for alerting that aids in offering right decision at right time. At first, remote sensing image data are collected and preprocessed using Frequency Ratio and Multi-collinearity test (MCT) to ensure noise removal and image augmentation by enhancing their quality. A feature extraction process is carried with the use of Deep Convolution VGGNet-16 from which optimal features are selected using Improved Harris Hawks Optimization algorithm (IHHOA). Then, a Flexible Spatiotemporal image fusion (F-SPTF) approach is employed to fuse images. After this, Deep cascaded RNN classifier is employed for predicting flood occurrence and to map flood susceptibility areas. This, in turn, classifies the normal and abnormal condition of flood occurrence thus giving alerts in case of natural disaster occurrences which could be visualized through digital twin technologies. At last, the performance is carried for various metrics like F1-score, precision, accuracy, recall, RMSE, NSE, MAPE, and MAE to validate proposed scheme efficiency. A suggested scheme offers an accuracy rate of about (99.89%), precision (99.37%), recall (99.82%), and F1-score (99.74%). The error rates estimated like RMSE (0.784), MAE (0.764), and MAPE (0.102) also seems to be lower th

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Automated face recognition using deep learning technique and center symmetric multivariant local binary pattern

引用

Neural Computing and Applications 2025年第1期37卷 263-281页

作者： Sekhar, J.C. Josephson, P. Joel Chinnasamy, A. Maheswari, M. Sankar, S. Kalangi, Ruth Ramya Department of Computer Science and Engineering NRI Institute of Technology Andhra Pradesh Guntur India Department of Electronics and Communication Engineering Malla Reddy Engineering College Telangana Hyderabad India Department of Data Science and Business Systems School of Computing SRMIST Kattankulathur Chennai India Department of Computer Science and Engineering Panimalar Engineering College Chennai India Department of Computer Science and Engineering Saveetha School of Engineering SIMATS Chennai India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Andhra Pradesh Vijayawada India

Researchers have recently created several deep learning strategies for various tasks, and facial recognition has made remarkable progress in employing these techniques. Face recognition is a noncontact, nonobligatory, acceptable, and harmonious biometric recognition method with a promising national and social security future. The purpose of this paper is to improve the existing face recognition algorithm, investigate extensive data-driven face recognition methods, and propose a unique automated face recognition methodology based on generative adversarial networks (GANs) and the center symmetric multivariable local binary pattern (CS-MLBP). To begin, this paper employs the center symmetric multivariant local binary pattern (CS-MLBP) algorithm to extract the texture features of the face, addressing the issue that C2DPCA (column-based two-dimensional principle component analysis) does an excellent job of removing the global characteristics of the face but struggles to process the local features of the face under large samples. The extracted texture features are combined with the international features retrieved using C2DPCA to generate a multifeatured face. The proposed method, GAN-CS-MLBP, syndicates the power of GAN with the robustness of CS-MLBP, resulting in an accurate and efficient face recognition system. Deep learning algorithms, mainly neural networks, automatically extract discriminative properties from facial images. The learned features capture low-level information and high-level meanings, permitting the model to distinguish among dissimilar persons more successfully. To assess the proposed technique’s GAN-CS-MLBP performance, extensive experiments are performed on benchmark face recognition datasets such as LFW, YTF, and CASIA-WebFace. Giving to the findings, our method exceeds state-of-the-art facial recognition systems in terms of recognition accuracy and resilience. The proposed automatic face recognition system GAN-CS-MLBP provides a solid basis for a

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Breaking Through the Spike: Spike Window Decoding for Accelerated and Precise Automatic Speech Recognition

Breaking Through the Spike: Spike Window Decoding for Accele...

引用

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

作者： Zhang, Wei Zhang, Tian-Hao Luo, Chao Zhou, Hui Yang, Chao Qian, Xinyuan Yin, Xu-Cheng *** GROUP Shanghai China School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China

ISBN: (纸本)9798350368741

Recently, end-to-end automatic speech recognition has become the mainstream approach in both industry and academia. To optimize system performance in specific scenarios, the Weighted Finite-State Transducer (WFST) is extensively used to integrate acoustic and language models, leveraging its capacity to implicitly fuse language models within static graphs, thereby ensuring robust recognition while also facilitating rapid error correction. However, WFST necessitates a frame-by-frame search of CTC posterior probabilities through autoregression, which significantly hampers inference speed. In this work, we thoroughly investigate the spike property of CTC outputs and further propose the conjecture that adjacent frames to non-blank spikes carry semantic information beneficial to the model. Building on this, we propose the Spike Window Decoding algorithm, which greatly improves the inference speed by making the number of frames decoded in WFST linearly related to the number of spiking frames in the CTC output, while guaranteeing the recognition performance. Our method achieves SOTA recognition accuracy with significantly accelerates decoding speed, proven across both AISHELL-1 and large-scale In-House datasets, establishing a pioneering approach for integrating CTC output with WFST. © 2025 IEEE.

关键词： automatic speech recognition spike window decoding wfst

来源：评论

学校读者我要写书评

暂无评论

DESIGN OF RF FEEDING NETWORK [1:8] USING UNEQUAL AMPLITUDE HYBRID COUPLER FOR LOW SLL X-BAND RADAR APPLICATIONS

引用

Telecommunications and Radio engineering (English translation of Elektrosvyaz and Radiotekhnika) 2025年第5期84卷 1-19页

作者： Kothapudi, Venkata Kishore Sunayana, Lella Divya, Shaik Kumar, Vijay Pappula, Lakshman Gottapu, Srinivasa Kiran Guntur District Andhra Pradesh Vadlamudi522213 India Department of Communication Engineering School of Electronics Engineering Vellore Institute of Technology Tamil Nadu Vellore632014 India Department of Electrical Electronics and Communication Engineering GITAM School of Technology GITAM University Andhra Pradesh Visakhapatnam530045 India School of Science and Engineering Engineering and Computer Science Department Benedict College 1600 Harden St. ColumbiaSC29204 United States

The design simulation and manufacturing of an x-band frequency uneven amplitude 90° hybrid coupler are described in this paper. This hybrid coupler is used to create a feeder network with eight output ports operating at 9.3 GHz. The overall goal is to design and analyze an uneven power feeding network (1:8) for low SLL x-band radar applications employing a hybrid coupler. The specific goal is to create a hybrid coupler with equal amplitude, a hybrid coupler with uneven amplitude [Taylor’s amplitude = –17 dB], and a feeding network using hybrid couplers. The goal of this research is to improve the ability of a single transmitter to provide power to more than one antenna, all at a time either equally or unequally, and this can be achieved by designing a desired number of hybrid couplers and connecting them to the specific number of antennas at the output end and to the single transmitter at the input end. © 2025 by Begell House, Inc.

关键词： Transmitters

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：