检索结果-内蒙古大学图书馆

24th COTA International Conference of Transportation Professionals: Resilient, Intelligent, Connected, and Lowcarbon Multimodal Transportation, CICTP 2024

作者： Wang, Congyu Wang, Chong Li, Yueqi Ye, Guoqiang Bao, Licheng Wen, Ming Zhang, Jian School of Software Nanjing Univ. of Information Science and Technology Nanjing China School of Computer Science Nanjing Univ. of Information Science and Technology Nanjing China School of Transportation Southeast Univ. Nanjing China

ISBN: (数字)9780784485484

ISBN: (纸本)9780784485484

Intelligent traffic signal control plays a crucial role in reducing the escalating problem of traffic congestion. However, traditional methods of traffic signal control struggle to effectively adapt to the ever-changing road conditions and demands in complex urban traffic environments. This becomes particularly problematic when considering the punctuality of buses, as it can lead to traffic congestion for other vehicles. To address this issue, this paper introduces a multi-agent reinforcement learning method that prioritizes bus punctuality while minimizing travel time for regular vehicles. We established the SUMO simulation environment with dedicated bus lanes and designed a reward function based on bus arrival and vehicle waiting times. By employing a deep reinforcement learning approach, we aimed to reduce average travel time. The simulation results demonstrate significant improvements, including a 74.44% reduction in vehicle waiting time at intersections and a 78.06% decrease in bus delay time. Notably, even when focusing solely on optimizing vehicle travel time, our approach enhances bus punctuality, resulting in a 57.03% reduction in bus delay time. Thus, we believe that our approach offers a viable solution for addressing bus priority policy challenges. © ASCE.

关键词： Travel time

来源：评论

学校读者我要写书评

暂无评论

Heart-Rate Analysis of Healthy and Insomnia Groups with Detrended Fractal Dimension Feature in Edge

引用

Tsinghua science and technology 2022年第2期27卷 325-332页

作者： Xuefei Wang Yichao Zhou Chunxia Zhao the School of Computer Science and Engineering Nanjing University of Science and TechnologyNanjing 210094China

Insomnia,whether situational or chronic,affects over a third of the general population in today’s ***,given the lack of non-contact and non-inductive quantitative evaluation approaches,most insomniacs are often unrecognized and *** Polysomnographic(PSG)is considered as one of the assessment methods,it is poorly tolerated and *** this paper,with the recent development of Internet-of-Things devices and edge computing techniques,we propose a detrended fractal dimension(DFD)feature for the analysis of heart-rate signals,which can be easily acquired by many wearables,of good sleepers and *** feature was derived by calculating the fractal dimension(FD)of detrended *** the trend component removal,we improved the null space pursuit algorithm and proposed an adaptive trend extraction *** experimental results demonstrated the efficacy of the proposed DFD index through numerical statistics and significance testing for healthy and insomnia groups,which renders it a potential biomarker for insomnia assessment and management.

关键词： insomnia fractal dimension adaptive signal separation hypothesis testing

来源：评论

学校读者我要写书评

暂无评论

High-speed encrypted traffic classification by using payload features

引用

Digital Communications and Networks 2025年第2期11卷 412-423页

作者： Xinge Yan Liukun He Yifan Xu Jiuxin Cao Liangmin Wang Guyang Xie School of Cyber Science and Engineering Southeast UniversityNanjing 211189China Nanjing Huafei Data Technology Co. Ltd.Nanjing 210019China School of Computer Science and Communication Engineering Jiangsu UniversityZhenjiang 212013China

Traffic encryption techniques facilitate cyberattackers to hide their presence and *** classification is an important method to prevent network ***,due to the tremendous traffic volume and limitations of computing,most existing traffic classification techniques are inapplicable to the high-speed network *** this paper,we propose a High-speed Encrypted Traffic Classification(HETC)method containing two ***,to efficiently detect whether traffic is encrypted,HETC focuses on randomly sampled short flows and extracts aggregation entropies with chi-square test features to measure the different patterns of the byte composition and distribution between encrypted and unencrypted ***,HETC introduces binary features upon the previous features and performs fine-grained traffic classification by combining these payload features with a Random Forest *** experimental results show that HETC can achieve a 94%F-measure in detecting encrypted flows and a 85%–93%F-measure in classifying fine-grained flows for a 1-KB flow-length dataset,outperforming the state-of-the-art comparison ***,HETC does not need to wait for the end of the flow and can extract mass computing *** average time for HETC to process each flow is only 2 or 16 ms,which is lower than the flow duration in most cases,making it a good candidate for high-speed traffic classification.

关键词： Traffic classification Flow analysis Information entropy Machine learning Randomness test

来源：评论

学校读者我要写书评

暂无评论

A Rearrangement and Restore-Based Mixer Model for Target-Oriented Multimodal Sentiment Classification

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第6期5卷 3109-3119页

作者： Jia, Li Ma, Tinghuai Rong, Huan Sheng, Victor S. Huang, Xuejian Xie, Xintong Nanjing University of Information Science and Technology School of Computer Science Nanjing Jiangsu210044 China Nanjing University of Information Science and Technology School of Computer and Software Nanjing Jiangsu210044 China Jiangsu Ocean University School of Computer Engineering Lianyungang Jiangsu222005 China Nanjing University of Information Science and Technology School of Artificial Intelligence Nanjing Jiangsu210044 China Texas Tech University Department of Computer Science LubbockTX79409 United States

With the development of fine-grained multimodal sentiment analysis tasks, target-oriented multimodal sentiment (TMSC) analysis has received more attention, which aims to classify the sentiment of target with the help of textual and associated image features. Existing methods focus on exploring fine-grained image features and incorporate transformer-based complex fusion strategies, while ignoring the heavy computational burden. Recently, some lightweight multilayer perceptrons (MLP)-based methods have been successfully applied to multimodal sentiment classification tasks. In this article, we propose an effective rearrangement and restore mixer model (RR-Mixer) for TMSC, which dedicates the interaction of image, text, and targets along the modal-axis, sequential-axis, and feature channel-axis through rearrangement and restore operations. Specifically, we take vision transformer (ViT) and robustly optimized BERT (RoBERTa) pretrained models to extract image and textual features, respectively. Further, we adopt cosine similarity to select the most semantically relevant image features. Then, an RR-Mixer module is designed for mixed multimodal features, with the core technology consisting of rolling, grouping rearrangement and restore operations. Moreover, we introduce MLP Unit to learn the information of different modalities for intermodal interaction. The results show that our model achieves superior performance on two benchmark multimodal datasets, TWITTER-15 and TWITTER-17, with a significant improvement of 4.66% and 1.26% in terms of macro-F1. © 2020 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

A Lyapunov Optimization Approach for Long-Term Task Assignment in Vehicular Crowdsensing

引用

IEEE Transactions on Consumer Electronics 2024年第1期71卷 2107-2117页

作者： Guo, Xuhui Wang, Xiumin Pan, Weijian Wu, Weiwei Liu, Kai South China University of Technology School of Computer Science and Engineering Guangzhou China Southeast University School of Computer Science and Engineering Nanjing China Chongqing University College of Computer Science Chongqing400030 China

Vehicular crowdsensing has recently received considerable attention, due to its promising capability of collecting useful information for the Internet of Vehicles. However, existing researches in crowdsensing mainly focus on optimizing the short-term objectives, e.g., quality of crowdsensing, social welfare, or crowdsensing cost. Few of them consider the assignment of crowdsensing tasks from a long-term perspective, particularly for tasks located within a sensing hole. Due to the lack of sufficient participants nearby, these tasks may not be timely completed and have to be delayed. Particularly, in a dynamic crowdsensing system, the requirements of tasks may also change over time, which presents a new challenge in achieving the system stability. To tackle these issues, this paper formulates a novel task assignment problem from a long term perspective. A Lyapunov optimization model is constructed to optimize both the stability of the system and the utility of crowdsensing. Additionally, we establish the upper bound of the utility function, based on which we construct an auxiliary graph to convert the defined task assignment problem into finding a minimum weight maximum flow from the graph. Finally, we perform simulations to validate the efficiency of the proposed scheme. © 1975-2011 IEEE.

关键词： Crowdsourcing

来源：评论

学校读者我要写书评

暂无评论

A Wavelet-based Diffusion Model for Low-light Image Enhancement in Raw Space 16

A Wavelet-based Diffusion Model for Low-light Image Enhancem...

引用

2024 16th International Conference on Graphics and Image Processing, ICGIP 2024

作者： Zhang, Xin-Yu Xu, Yang Wei, Zhi-Hui Wu, Ze-Bin School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

ISBN: (数字)9781510688780

ISBN: (纸本)9781510688773

Low-light image enhancement (LLIE) in Raw space has posed a challenge in the field of image processing and computational photography. Different from image processing in sRGB space, Raw images store more image information and thus have more space for image processing. Existing methods often struggle to approximate real images in terms of image detail enhancement and color restoration. Drawing inspiration from the latest advancements in denoising diffusion models, we introduce an innovative framework that leverages wavelet transform and diffusion models to tackle the low light image enhancement challenge in Raw space. In our method, the input low-light image is first decomposed into different frequency sub-bands by discrete wavelet transform (DWT) from both image and feature levels. Then a redesigned diffusion model is trained to adaptively denoise and enhance the different frequency sub-bands. In order to achieve better results in human eye perception, we also improved the loss function during training process. Finally, the enhanced result of the low-light image is finally reconstructed through inverse wavelet transform (IWT) and subsequent processing, transitioning from Raw space to sRGB space. Extensive experiments on SID and ELD datasets show that our refined method outperforms several existing methods, delivering superior performance both quantitatively and qualitatively. © 2025 SPIE.

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Multi-modal Knowledge Distillation-guided Artery-Vein Segmentation Network for OCT/OCTA Images 16

Multi-modal Knowledge Distillation-guided Artery-Vein Segmen...

引用

2024 16th International Conference on Graphics and Image Processing, ICGIP 2024

作者： Li, Shuo Zhang, Wenbo Chen, Qiang School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

ISBN: (数字)9781510688780

ISBN: (纸本)9781510688773

The analysis of artery and vein differences in Optical Coherence Tomography Angiography (OCTA) is of great significance for diagnosing various eye diseases and systemic diseases (such as diabetic retinopathy, hypertension, and cardiovascular diseases). We propose a multi-modal knowledge distillation-guided Artery-Vein Segmentation Network for OCT/OCTA images (MKD-Net). Firstly, we introduce 2D projection image information in the process of segmenting from 3D to 2D to enhance the semantic representation capability in the horizontal direction of the network. Secondly, we transfer knowledge from a trained multi-modal (3D and 2D input) teacher network to a single-modal (3D input) student network, enabling the student network to not rely on the projection images generated by retinal layer segmentation, avoiding segmentation failure under diseased conditions, and enhancing its ability to capture details and feature changes in the images. Extensive evaluations on the OCTA500 and OCTA100 datasets demonstrate that the proposed method achieves superior performance, with an MIou of 84.96% and 84.25% on the OCTA500 and OCTA100 datasets respectively. © 2025 SPIE.

关键词： Optical coherence tomography

来源：评论

学校读者我要写书评

暂无评论

Hybrid Query of Boolean Filter and Vector Similarity Search: Benchmark, Comparison and Direction 5

Hybrid Query of Boolean Filter and Vector Similarity Search:...

引用

5th Asia Conference on Information engineering, ACIE 2025

作者： Zhu, Zijian Wang, Yongli Liu, Dongmei School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

ISBN: (纸本)9798331528409

The growing use of vectors for unstructured data has made efficient hybrid queries-combining boolean filters with vector similarity searches-essential. However, publicly available datasets for evaluating DBMS performance on such queries are limited. To address this, we introduce two datasets: the singletable fungi dataset (STFD) with 295,938 base entities and the multi-table movie dataset (MTMD) with 284,713 base vectors, available at this link. These datasets provide a framework for evaluating DBMS performance on hybrid queries. Our experiments with Milvus, pgvector, ClickHouse, and Elasticsearch show that current DBMSs struggle with hybrid queries, each system exhibiting strengths and weaknesses. QPS varies up to 50 ×, and recall differences reach 55 %. These results underscore the importance of query optimization and index utilization. We outline future research directions, including specialized data structures, algorithms for hybrid queries, techniques for estimating query difficulty and predicate selectivity, and enhanced query optimization strategies, to support the development of more efficient DBMS solutions. © 2025 IEEE.

关键词： Structured Query Language

来源：评论

学校读者我要写书评

暂无评论

QRALadder: QoE and Resource Consumption-Aware Encoding Ladder Optimization for Live Video Streaming 31st

QRALadder: QoE and Resource Consumption-Aware Encoding Ladd...

引用

31st International Conference on Multimedia Modeling, MMM 2025

作者： Zhu, Yingqian Gao, Guanyu School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

ISBN: (纸本)9789819620630

Live video streaming demands high user Quality of Experience (QoE) and requires significant computing power and bandwidth for video encoding and transmission. The standard adaptive live streaming approach encodes the source video content at multiple predefined bitrates, allowing users to select the most appropriate bitrate for optimal rate adaptation. The current methods for encoding ladder design primarily focus on optimizing user QoE. However, the relationship between the encoding ladder and user QoE is complex due to the dynamic nature of live streaming, where the number of users fluctuates, and each user experiences varying and evolving bandwidth conditions. On the other hand, resource consumption in live streaming significantly impacts operational costs, as it requires substantial resources for both video encoding and delivery. To address this challenge, we propose QRALadder, a QoE and resource consumption-aware approach to encoding ladder design. Our method simultaneously optimizes user QoE while considering the impact of the encoding ladder on resource consumption in live streaming, and it can dynamically select the most suitable encoding ladder based on the current system state. We use the Dueling Double Deep Q Networks (Dueling DDQN) algorithm to learn the optimal encoding ladder, maximizing overall performance by analyzing user QoE feedback and resource consumption about system states. We conduct comparative experiments using various types of video content and real-world bandwidth traces to validate our approach. The results show that QRALadder can achieve higher overall performance compared to baseline methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Q-learning based strategy analysis of cyber-physical systems considering unequal cost

引用

Intelligent and Converged Networks 2023年第2期4卷 116-126页

作者： Xin Chen Jixiang Cheng Luanjuan Jiang Qianmu Li Ting Wang Dafang Li School of Management Science and Engineering Nanjing University of Finance and EconomicsNanjing 210023China School of Computer Science and Engineering Nanjing University of Science and TechnologyNanjing 210014China

This paper proposes a cyber security strategy for cyber-physical systems(CPS)based on Q-learning under unequal cost to obtain a more efficient and low-cost cyber security defense strategy with misclassification *** system loss caused by strategy selection errors in the cyber security of CPS is often considered ***,sometimes the cost associated with different errors in strategy selection may not always be the same due to the severity of the consequences of ***,unequal costs referring to the fact that different strategy selection errors may result in different levels of system losses can significantly affect the overall performance of the strategy selection *** introducing a weight parameter that adjusts the unequal cost associated with different types of misclassification errors,a modified Q-learning algorithm is proposed to develop a defense strategy that minimizes system loss in CPS with misclassification interference,and the objective of the algorithm is shifted towards minimizing the overall ***,simulations are conducted to compare the proposed approach with the standard Q-learning based cyber security strategy method,which assumes equal costs for all types of misclassification *** results demonstrate the effectiveness and feasibility of the proposed research.

关键词： cyber security Q-learning policy selection unequal cost misclassification interference

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：