检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Qing Liu Qian Gao Jun Fan Zhiqiang Zhang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Industrial Network and Information System Security Shandong Fundamental Research Center for Computer Science Jinan China China Telecom Digital Intelligence Techonology Co Ltd Jinan China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

In recent years, research has illuminated the potency of implicit data processing in enhancing user preferences. Nevertheless, barriers remain in breaking through the constraints of implicit information. This study aims to bridge this gap by firstly constructing a triangular stable node network model, tailored to manage implicit information with precision. Recognizing the challenge of pinpointing novel structures within large-scale graphs, we propose SLTSNN— a triangular stable node network based on self-supervised learning for personalized prediction. SLTSNN innovates by maximizing mutual information between graph-level and patch-level representations, while augmenting graph representations through extended enhanced representations. Additionally, it incorporates triangular data associations and introduces a triangular allocation attention network, which emphasizes strongly correlated preference features among similar users. Furthermore, SLTSNN employs contrastive learning to maximize mutual information between graph vectors and hidden representations, distinguishing between high-order global and local representations. The model’s effectiveness in enhancing user preferences and capturing new graph structures is evidenced by its performance on hit rate and normalized discounted cumulative gain metrics across three datasets.

关键词： Measurement Bridges Contrastive learning Signal processing data processing Vectors Space exploration Resource management Speech processing Mutual information

来源：评论

学校读者我要写书评

暂无评论

Advancing cancer diagnosis and prognostication through deep learning mastery in breast, colon, and lung histopathology with ResoMergeNet

引用

Computers in Biology and Medicine 2025年 185卷 109494-109494页

作者： Ejiyi, Chukwuebuka Joseph Qin, Zhen Agbesi, Victor K. Yi, Ding Atwereboannah, Abena A. Chikwendu, Ijeoma A. Bamisile, Oluwatoyosi F. Bakanina Kissanga, Grace-Mercure Bamisile, Olusola O. College of Nuclear Technology and Automation Engineering & Sichuan Industrial Internet Intelligent Monitoring and Application Engineering Research Center Chengdu University of Technology Sichuan Chengdu China Network and Data Security Key Laboratory of Sichuan Province University of Electronic Science and Technology of China Sichuan Chengdu China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu China School of Life Science University of Electronic Science and Technology of China Chengdu China

Cancer, a global health threat, demands effective diagnostic solutions to combat its impact on public health, particularly for breast, colon, and lung cancers. Early and accurate diagnosis is essential for successful treatment, prompting the rise of Computer-Aided Diagnosis Systems as reliable and cost-effective tools. Histopathology, renowned for its precision in cancer imaging, has become pivotal in the diagnostic landscape of breast, colon, and lung cancers. However, while deep learning models have been widely explored in this domain, they often face challenges in generalizing to diverse clinical settings and in efficiently capturing both local and global feature representations, particularly for multi-class tasks. This underscores the need for models that can reduce biases, improve diagnostic accuracy, and minimize error susceptibility in cancer classification tasks. To this end, we introduce ResoMergeNet (RMN), an advanced deep-learning model designed for both multi-class and binary cancer classification using histopathological images of breast, colon, and lung. ResoMergeNet integrates the Resboost mechanism which enhances feature representation, and the ConvmergeNet mechanism which optimizes feature extraction, leading to improved diagnostic accuracy. Comparative evaluations against state-of-the-art models show ResoMergeNet's superior performance. Validated on the LC-25000 and BreakHis (400× and 40× magnifications) datasets, ResoMergeNet demonstrates outstanding performance, achieving perfect scores of 100 % in accuracy, sensitivity, precision, and F1 score for binary classification. For multi-class classification with five classes from the LC25000 dataset, it maintains an impressive 99.96 % across all performance metrics. When applied to the BreakHis dataset, ResoMergeNet achieved 99.87 % accuracy, 99.75 % sensitivity, 99.78 % precision, and 99.77 % F1 score at 400× magnification. At 40× magnification, it still delivered robust results with 98.85 % accuracy,

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

PrintListener: Uncovering the Vulnerability of Fingerprint Authentication via the Finger Friction Sound

arXiv

引用

arXiv 2024年

作者： Zhou, Man Su, Shuao Wang, Qian Li, Qi Zhou, Yuting Ma, Xiaojing Li, Zhengxiong School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Cyber Science and Engineering Wuhan University China Institute for Network Sciences and Cyberspace Tsinghua University China Department of Computer Science and Engineering University of Colorado Denver United States Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering Huazhong University of Science and Technology China

Fingerprint authentication has been extensively employed in contemporary identity verification systems owing to its rapidity and cost-effectiveness. Due to its widespread use, fingerprint leakage may cause sensitive information theft, enormous economic and personnel losses, and even a potential compromise of national security. As a fingerprint that can coincidentally match a specific proportion of the overall fingerprint population, MasterPrint rings the alarm bells for the security of fingerprint authentication. In this paper, we propose a new side-channel attack on the minutiae-based Automatic Fingerprint Identification System (AFIS), called PrintListener, which leverages users’ fingertip swiping actions on the screen to extract fingerprint pattern features (the first-level features) and synthesizes a stronger targeted PatternMasterPrint with potential second-level features. The attack scenario of PrintListener is extensive and covert. It only needs to record users’ fingertip friction sound and can be launched by leveraging a large number of social media platforms. Extensive experimental results in real-world scenarios show that Printlistener can significantly improve the attack potency of MasterPrint. © 2024, CC0.

关键词： Friction

来源：评论

学校读者我要写书评

暂无评论

A Survey of Attacks on Large Vision-Language Models: Resources, Advances, and Future Trends

arXiv

引用

arXiv 2024年

作者： Liu, Daizong Yang, Mingyu Qu, Xiaoye Zhou, Pan Cheng, Yu Hu, Wei Wangxuan Institute of Computer Technology Peking University No. 128 Zhongguancun North Street Beijing China The Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan430074 China The Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong

With the significant development of large models in recent years, Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities across a wide range of multimodal understanding and reasoning tasks. Compared to traditional Large Language Models (LLMs), LVLMs present great potential and challenges due to its closer proximity to the multi-resource real-world applications and the complexity of multi-modal processing. However, the vulnerability of LVLMs is relatively underexplored, posing potential security risks in daily usage. In this paper, we provide a comprehensive review of the various forms of existing LVLM attacks. Specifically, we first introduce the background of attacks targeting LVLMs, including the attack preliminary, attack challenges, and attack resources. Then, we systematically review the development of LVLM attack methods, such as adversarial attacks that manipulate model outputs, jailbreak attacks that exploit model vulnerabilities for unauthorized actions, prompt injection attacks that engineer the prompt type and pattern, and data poisoning that affects model training. Finally, we discuss promising research directions in the future. We believe that our survey provides insights into the current landscape of LVLM vulnerabilities, inspiring more researchers to explore and mitigate potential safety issues in LVLM developments. The latest papers on LVLM attacks are continuously collected in https://***/liudaizong/Awesome-LVLM-Attack. Copyright © 2024, The Authors. All rights reserved.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

VG-Net: Sensor Time Series Anomaly Detection with Joint Variational Autoencoder and Graph Neural Network

VG-Net: Sensor Time Series Anomaly Detection with Joint Vari...

引用

Smart World Congress (SWC), IEEE

作者： Shuai Wang Shenghao Liu Xiaoxuan Fan Haijun Wang Yinxin Zhou Hanjun Gao Hongwei Lu Xianjun Deng Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China Hunan Technical College of Railway High-speed China Nanjing University of Posts and Telecommunications China China National Nuclear Power Co Ltd China

ISBN: (数字)9798331520861

ISBN: (纸本)9798331520878

Intelligent Cyber-Physical Systems typically utilize sensors to gather a significant amount of raw data which often possesses temporal and high-dimensional characteristics. Ensuring the data reliability is crucial for the normal operation of the system. While most existing methods are capable of identifying anomalies in high-dimensional time series data from sensors, they often lack adequate accuracy and fail to adequately consider the correlation between sensor nodes. To address these limitations, we propose an unsupervised framework which is a combination of variational autoencoder and graph neural network(VG-Net). By jointly optimizing the reconstruction module and the prediction module, VG-Net effectively models the correlation between sensor nodes while simultaneously detecting anomalies. The reconstruction module is responsible for modeling the latent representation space and reconstructing the time series data through variational autoencoder. The prediction module leverages bidirectional long short term memory networks and neural tensor network to learn the relationships between sensor nodes, and uses graph neural network to predict the future behavior of the data. Experiments on eight real and synthetic datasets demonstrate that, compared with existing methods, VG-Net can accurately model the relationship between sensor nodes and enhance the anomaly detection performance of high-dimensional time series data.

关键词： Time series analysis Autoencoders Sensor phenomena and characterization Predictive models data models Graph neural networks Reliability Anomaly detection Intelligent sensors Synthetic data

来源：评论

学校读者我要写书评

暂无评论

A method for detecting floating objects on water based on edge computing

A method for detecting floating objects on water based on ed...

引用

IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)

作者： He Li Shuaipeng Yang Jinjiang Liu Honglin Fang Zhumu Fu Rui Zhang Huimei Jia Lianmeng Lv Henan Costar Group Co. Ltd Nanyang Henan China College of Information Engineering Henan University of Science and Technology Luoyang Henan China Henan Engineering Research Center of Intelligent Processing for Big Data of Digital Image School of Computer Science and Technology Nanyang Normal University Nanyang China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunication Beijing China Xi'an Hengpin Electronic Technology Co. Ltd Xi’an China

With the development and application of computer vision, many target detection networks are applied to the detection of floating objects in rivers. For the detection problems such as small targets easily missed and misdetected in water surface floating object detection tasks and difficult to deploy models. An edge computing-oriented approach to river floater detection is proposed. First, a four-fold down-sampling feature layer is added to the YOLOv5 network which enhances more target detail features and improves the detection capability of small objects. Second, CA (Coordinate Attention) is added to the Backbone to suppress background noise interference, and different pooling is used to accommodate different hierarchical features. Then, a bilinear interpolation method is adopted for up-sampling to avoid the loss of small object features. Design a data enhancement algorithm for small targets based on Mosaic to increase the number of small objects and enrich the training background. Finally, for the edge computing architecture platform, the channel pruning algorithm is used to prune and compress the model structure to adapt to the computing capability of edge devices. The experimental results show that the method can effectively improve the detection capability of the network for floating objects on the water surface. The detection accuracy can reach 93.6%, and the detection speed can be maintained at 36 frames per second, which can achieve high-precision real-time detection of floating objects on the water surface.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MorphText: Deep Morphology Regularized Accurate Arbitrary-shape Scene Text Detection

arXiv

引用

arXiv 2024年

作者： Xu, Chengpei Jia, Wenjing Wang, Ruomei Luo, Xiaonan He, Xiangjian Faculty of Engineering and IT University of Technology Sydney Australia The Department of Computer Science University of Nottingham Ningbo China School of Data and Computer Science National Engineering Research Center of Digital Life Sun Yat-sen University China The School of Computer Science and Information Security Guilin University of Electronic Technology China

Bottom-up text detection methods play an important role in arbitrary-shape scene text detection but there are two restrictions preventing them from achieving their great potential, i.e., 1) the accumulation of false text segment detections, which affects subsequent processing, and 2) the difficulty of building reliable connections between text segments. Targeting these two problems, we propose a novel approach, named "MorphText", to capture the regularity of texts by embedding deep morphology for arbitrary-shape text detection. Towards this end, two deep morphological modules are designed to regularize text segments and determine the linkage between them. First, a Deep Morphological Opening (DMOP) module is constructed to remove false text segment detections generated in the feature extraction process. Then, a Deep Morphological Closing (DMCL) module is proposed to allow text instances of various shapes to stretch their morphology along their most significant orientation while deriving their connections. Extensive experiments conducted on four challenging benchmark datasets (CTW1500, Total-Text, MSRA-TD500 and ICDAR2017) demonstrate that our proposed MorphText outperforms both top-down and bottom-up state-of-the-art arbitrary-shape scene text detection approaches. Copyright © 2024, The Authors. All rights reserved.

关键词： Morphology

来源：评论

学校读者我要写书评

暂无评论

Downstream-agnostic Adversarial Examples

arXiv

引用

arXiv 2023年

作者： Zhou, Ziqi Hu, Shengshan Zhao, Ruizhi Wang, Qian Zhang, Leo Yu Hou, Junhui Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China School of Cyber Science and Engineering Wuhan University China School of Information and Communication Technology Griffith University Australia Department of Computer Science City University of Hong Kong Hong Kong National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security China Hubei Engineering Research Center on Big Data Security China Cluster and Grid Computing Lab

Self-supervised learning usually uses a large amount of unlabeled data to pre-train an encoder which can be used as a general-purpose feature extractor, such that downstream users only need to perform fine-tuning operations to enjoy the benefit of "large model". Despite this promising prospect, the security of pre-trained encoder has not been thoroughly investigated yet, especially when the pre-trained encoder is publicly available for commercial use. In this paper, we propose AdvEncoder, the first framework for generating downstream-agnostic universal adversarial examples based on the pre-trained encoder. AdvEncoder aims to construct a universal adversarial perturbation or patch for a set of natural images that can fool all the downstream tasks inheriting the victim pre-trained encoder. Unlike traditional adversarial example works, the pre-trained encoder only outputs feature vectors rather than classification labels. Therefore, we first exploit the high frequency component information of the image to guide the generation of adversarial examples. Then we design a generative attack framework to construct adversarial perturbations/patches by learning the distribution of the attack surrogate dataset to improve their attack success rates and transferability. Our results show that an attacker can successfully attack downstream tasks without knowing either the pre-training dataset or the downstream dataset. We also tailor four defenses for pre-trained encoders, the results of which further prove the attack ability of AdvEncoder. Our codes are available at: https://github. com/CGCL-codes/AdvEncoder. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Optimal Dispatch for Operating Reserve Considering Impact of Multi-Dimensional Prediction Information on Prediction Error

Optimal Dispatch for Operating Reserve Considering Impact of...

引用

Frontiers of Power and Energy (ACFPE), Asian Conference on

作者： Yang He Huixuan Li Shuo Yin Sheng Liu Mingxu Xiang Henan Power Trading Center Co. Ltd Zhengzhou Henan China State Grid Henan Electric Power Company Economic and Technological Research Institute Zhengzhou Henan China State Key Laboratory of Power Transmission Equipment & System Security and New Technology College of Electrical Engineering Chongqing University Chongqing China

ISBN: (数字)9798350303896

ISBN: (纸本)9798350303902

The large-scale integration of renewables has become an inevitable trend in China toward the target of “carbon peak and carbon neutrality”. Strong uncertainties of renewables will bring great challenges for the security operation of power systems. The reasonable dispatch of the operating reserve is an important way to handle the uncertainties. The stochastic dispatch method can effectively consider the uncertainties. However, this method faces computational obstacles. Therefore, the deterministic dispatch method is used in industrial practices. The key point is to properly determine the operating reserve requirement before the dispatch. However, the existing method does not consider the impact of multidimensional prediction information on the prediction error of net load. As a result, the determined operating reserve may be conservative or even insufficient. In view of the above problem, based on the deterministic dispatch framework, an optimal dispatch method for operating reserve considering the impact of multi-dimensional prediction information on prediction error is proposed in this paper. Firstly, the distribution of the system overall prediction error is estimated considering the impact of multi-dimensional prediction information, including meteorology and predicted power. Then, with the given confidence level, the operating reserve requirement is quantified. Finally, the total operating reserve requirement is allocated to each unit through the deterministic power dispatch. Case studies based on practical data validate the effectiveness of the proposed method.

关键词： Renewable energy sources Uncertainty Carbon neutral Market research Large scale integration Power systems security

来源：评论

学校读者我要写书评

暂无评论

DRLFcc: Deep Reinforcement Learning-empowered Congestion Control Mechanism for TCP Fast Recovery in High Loss Wireless Networks

DRLFcc: Deep Reinforcement Learning-empowered Congestion Con...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Yuanlong Cao Jinquan Nie Yuehua Fan Xun Shao Gang Lei School of Software Jiangxi Normal University Nanchang China Jiangxi Provincial Engineering Research Center of Blockchain Data Security and Governance Nanchang China Management Science and Engineering Center Jiangxi Normal University Nanchang China Toyohashi University of Technology Toyohashi Japan

TCP is currently the most widely used Internet transmission protocol, which is extensively applied to applications on the Internet to enable reliable data transmission. The TCP congestion control algorithm has a significant performance impact on all applications that use the TCP. However, traditional TCP congestion control algorithms rely on fixed feedback mechanisms, which can be challenging to adapt to complex and changing network environments and application scenarios, resulting in network performance bottlenecks. To address this issue, we design a congestion control windowing solution, DRLFcc, which is based on deep reinforcement learning and the TCP fast recovery mechanism. DRLFcc has demonstrated the ability to facilitate real-time adaptation of the congestion window to dynamic changes in network conditions while incorporating fast recovery mechanisms, thereby effectively enhancing network throughput and improving data transmission capacity recovery in high-loss wireless networks. The DRLFcc algorithm is validated in NS-3, and experimental results show an average improvement of 196% in effective throughput and a 22.4% reduction in round trip time. Compared to traditional TCP congestion control algorithms, the DRLFcc algorithm demonstrates superior performance and robustness.

关键词：

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：