检索结果-内蒙古大学图书馆

Diagnosis of COVID-19 Pneumonia via a Novel Deep Learning architecture

Journal of computer science & Technology 2022年第2期37卷 330-343页

作者： Xin Zhang Siyuan Lu Shui-Hua Wang Xiang Yu Su-Jing Wang Lun Yao Yi Pan Yu-Dong Zhang Department of Medical Imaging The Fourth People's Hospital of Huai'anHuai'an 223002China School of Informatics University of LeicesterLeicesterLE17RHU.K. School of Architecture Building and Civil Engineering Loughborough UniversityLoughboroughLE113TUU.K. School of Mathematics and Actuarial Science University of LeicesterLeicesterLE17RHU.K. Key Laboratory of Behavior Sciences Institute of PsychologyChinese Academy of SciencesBeijing 100101China Department of Psychology University of the Chinese Academy of SciencesBeijing 100101China Department of Infection Diseases The Fourth People's Hospital of Huai'anHuaian 223002China Department of Computer Science Georgia State UniversityAtlanta 30302-5060U.S.A. Department of Information Systems Faculty of Computing and Information TechnologyKing Abdulaziz University Jeddah 21589Saudi Arabia

COVID-19 is a contagious infection that has severe effects on the global economy and our daily *** diagnosis of COVID-19 is of importance for consultants,patients,and *** this study,we use the deep learning network AlexNet as the backbone,and enhance it with the following two aspects:1)adding batch normalization to help accelerate the training,reducing the internal covariance shift;2)replacing the fully connected layer in AlexNet with three classifiers:SNN,ELM,and ***,we have three novel models from the deep COVID network(DC-Net)framework,which are named DC-Net-S,DC-Net-E,and DC-Net-R,*** comparison,we find the proposed DC-Net-R achieves an average accuracy of 90.91%on a private dataset(available upon email request)comprising of 296 images while the specificity reaches 96.13%,and has the best performance among all three proposed *** addition,we show that our DC-Net-R also performs much better than other existing algorithms in the literature.

关键词： pneumonia COVID-19 convolutional neural network AlexNet deep learning

来源：评论

学校读者我要写书评

暂无评论

Fast Event-based Double Integral for Real-time Robotics

Fast Event-based Double Integral for Real-time Robotics

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Shijie Lin Yingqiang Zhang Dongyue Huang Bin Zhou Xiaowei Luo Jia Pan Department of Computer Science The University of Hong Kong Hong Kong SAR China Peng Cheng Laboratory Shenzhen Guangdong China Department of Mechanical and Automation Engineering The Chinese University of Hong Kong Hong Kong SAR China School of Computer Science and Engineering Beihang University Beijing China Department of Architecture and Civil En-gineering City University of Hong Kong Hong Kong SAR China

Motion deblurring is a critical ill-posed problem that is important in many vision-based robotics applications. The recently proposed event-based double integral (EDI) provides a theoretical framework for solving the deblurring prob-lem with the event camera and generating clear images at high frame-rate. However, the original EDI is mainly designed for offline computation and does not support real-time requirement in many robotics applications. In this paper, we propose the fast EDI, an efficient implementation of EDI that can achieve real-time online computation on single-core CPU devices, which is common for physical robotic platforms used in practice. In experiments, our method can handle event rates at as high as 13 million event per second in a wide variety of challenging lighting conditions. We demonstrate the benefit on multiple downstream real-time applications, including localization, vi-sual tag detection, and feature matching.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Impact of Soil Layer Distribution on the Morphology Characteristics of Underground Cavities

引用

Journal of Performance of Constructed Facilities 2025年第4期39卷

作者： Cao, Liting Chen, Yan Li, Xinyue Li, Jiacheng Djamaluddine, Ibrahim Lv, Xiangfeng School of Civil and Resources Engineering Univ. of Science and Technology Beijing Beijing100083 China Engineering College Hasanuddin Univ. Makassar90245 Indonesia

Urban road collapse disasters pose a serious threat to public safety and national economic development. The underground cavities formed by drainage pipe exfiltration are the primary cause of road collapse disasters in beijing, with soil layer distributions playing a crucial role in their formation and development. In this study, a physical model test system for road collapse induced by drainage pipe exfiltration was developed, which comprised a drainage control unit, pressure regulation unit, experiment platform, multiple information monitoring unit, and water circulation unit. Internal and external imaging, coupled with waterproof laser rangefinders, were employed to study the formation, development, and collapse morphologies of the underground cavity influenced by clay, sand-clay, and clay-sand distributions. The results indicated that the underground cavity formed in four phases: soil-water mixture adhesion, seepage-induced runoff generation, irregular void development, and arched cavity formation. Furthermore, the physical and mechanical properties of the soil in proximity to the drainage pipe exerted the most significant influence on cavity development. The vertical cross section of the underground cavity was mushroom cloud-shaped, characterized by an upper bearing structure zone, middle saturated rheology zone, and lower stacking repose zone. The findings provide theoretical support to reveal the mechanisms of road collapse. © 2025 American Society of civil Engineers.

关键词： Seepage

来源：评论

学校读者我要写书评

暂无评论

Structure-preserving progressive low-rank image completion for defending adversarial attacks

arXiv

引用

arXiv 2021年

作者： Zhao, Zhiqun Wang, Hengyou Sun, Hao He, Zhihai Department of Electrical Engineering and Computer Science University of Missouri ColumbiaMO65203 United States School of Science Beijing University of Civil Engineering and Architecture Beijing100044 China

Deep neural networks recognize objects by analyzing local image details and summarizing their information along the inference layers to derive the final decision. Because of this, they are prone to adversarial attacks. Small sophisticated noise in the input images can accumulate along the network inference path and produce wrong decisions at the network output. On the other hand, human eyes recognize objects based on their global structure and semantic cues, instead of local image textures. Because of this, human eyes can still clearly recognize objects from images which have been heavily damaged by adversarial attacks. This leads to a very interesting approach for defending deep neural networks against adversarial attacks. In this work, we propose to develop a structure-preserving progressive low-rank image completion (SPLIC) method to remove unneeded texture details from the input images and shift the bias of deep neural networks towards global object structures and semantic cues. We formulate the problem into a low-rank matrix completion problem with progressively smoothed rank functions to avoid local minimums during the optimization process. Our experimental results demonstrate that the proposed method is able to successfully remove the insignificant local image details while preserving important global object structures. On black-box, gray-box, and white-box attacks, our method outperforms existing defense methods (by up to 12.6%) and significantly improves the adversarial robustness of the network. Copyright © 2021, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Sequence-to-Point Learning Based on Spatio-Temporal Attention Fusion Network for Non-Intrusive Load Monitoring

SSRN

引用

SSRN 2023年

作者： Zhang, Shiqing Wang, Lei Fu, Youyao Guo, Xianhai Zhao, Xiaoming Fang, Jiangxiong Zhang, Zhen Liu, Yadong Wang, Xiaoli Zhang, Baochang Yu, Jun Institute of Intelligent Information Processing Taizhou University Zhejiang Province Taizhou318000 China School of Civil Engineering and Architecture Taizhou University Zhejiang Province Taizhou318000 China Zhejiang Wellsun Intelligent Technology Company Limited Zhejiang Province Tiantai317200 China School of Automation Science and Electrical Engineering Beihang University Beijing100191 China School of Computer Science Hangzhou Dianzi University Hangzhou310018 China

Non-invasive load monitoring (NILM) aims to disaggregate the power consumption of single electrical appliance from the aggregated electrical appliances. Most of existing NILM methods ignores the complementarity between temporal and spatial characteristics of appliance power data. To tackle this problem, this paper proposes a spatio-temporal attention fusion network with a sequence-to-point learning scheme for load disaggregation. Initially, a temporal feature extraction module consisting of several bi-directional (non-causal) temporal convolution network residual blocks is designed to extract temporal features over a large temporal receptive field. Then, an asymmetric inception module is designed for a multi-scale spatial feature extraction. The extracted spatial feature and temporal features are concatenated, and fed into a polarized self-attention module to perform spatio-temporal attention fusion, followed by two dense layers for final NILM predictions. Extensive experiments on two public datasets such as REDD and UK-DALE show the validity of the proposed method, outperforming other used methods on NILM tasks. © 2023, The Authors. All rights reserved.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

SwinTrack: a simple and strong baseline for transformer tracking 22

SwinTrack: a simple and strong baseline for transformer trac...

引用

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Liting Lin Heng Fan Zhipeng Zhang Yong Xu Haibin Ling School of Computer Science & Engineering South China Univ. of Tech. Guangzhou China and Peng Cheng Laboratory Shenzhen China Department of Computer Science and Engineering University of North Texas Denton DiDi Chuxing Beijing China Department of Computer Science Stony Brook University Stony Brook

ISBN: (纸本)9781713871088

Recently Transformer has been largely explored in tracking and shown state-of-the-art (SOTA) performance. However, existing efforts mainly focus on fusing and enhancing features generated by convolutional neural networks (CNNs). The potential of Transformer in representation learning remains under-explored. In this paper, we aim to further unleash the power of Transformer by proposing a simple yet efficient fully-attentional tracker, dubbed SwinTrack, within classic Siamese framework. In particular, both representation learning and feature fusion in SwinTrack leverage the Transformer architecture, enabling better feature interactions for tracking than pure CNN or hybrid CNN-Transformer frameworks. Besides, to further enhance robustness, we present a novel motion token that embeds historical target trajectory to improve tracking by providing temporal context. Our motion token is lightweight with negligible computation but brings clear gains. In our thorough experiments, SwinTrack exceeds existing approaches on multiple benchmarks. Particularly, on the challenging LaSOT, SwinTrack sets a new record with 0.713 SUC score. It also achieves SOTA results on other benchmarks. We expect SwinTrack to serve as a solid baseline for Transformer tracking and facilitate future research.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Spatial and Spatial-Spectral Morphological Mamba for Hyperspectral Image Classification

arXiv

引用

arXiv 2024年

作者： Ahmad, Muhammad Butt, Muhammad Hassaan Farooq Khan, Adil Mehmood Mazzara, Manuel Distefano, Salvatore Usama, Muhammad Roy, Swalpa Kumar Chanussot, Jocelyn Hong, Danfeng Dipartimento di Matematica e Informatica-MIFT University of Messina Messina98121 Italy Institute of Artificial Intelligence School of Mechanical and Electrical Engineering Shaoxing University Shaoxing312000 China School of Computer Science University of Hull HullHU6 7RX United Kingdom Institute of Software Development and Engineering Innopolis University Innopolis420500 Russia Department of Computer Science National University of Computer and Emerging Sciences Islamabad Chiniot-Faisalabad Campus Chiniot35400 Pakistan Department of Computer Science and Engineering Alipurduar Government Engineering and Management College West Bengal 736206 India Univ. Grenoble Alpes Inria CNRS Grenoble INP LJK Grenoble38000 France The Aerospace Information Research Institute Chinese Academy of Sciences Beijing100094 China Aerospace Information Research Institute Chinese Academy of Sciences Beijing100094 China School of Electronic Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing100049 China

Recent advancements in transformers, specifically self-attention mechanisms, have significantly improved hyperspectral image (HSI) classification. However, these models often suffer from inefficiencies, as their computational complexity scales quadratically with sequence length. To address these challenges, we propose the morphological spatial mamba (SMM) and morphological spatial-spectral Mamba (SSMM) model (MorpMamba), which combines the strengths of morphological operations and the state space model framework, offering a more computationally efficient alternative to transformers. In MorpMamba, a novel token generation module first converts HSI patches into spatial-spectral tokens. These tokens are then processed through morphological operations such as erosion and dilation, utilizing depthwise separable convolutions to capture structural and shape information. A token enhancement module refines these features by dynamically adjusting the spatial and spectral tokens based on central HSI regions, ensuring effective feature fusion within each block. Subsequently, multi-head self-attention is applied to further enrich the feature representations, allowing the model to capture complex relationships and dependencies within the data. Finally, the enhanced tokens are fed into a state space module, which efficiently models the temporal evolution of the features for classification. Experimental results on widely used HSI datasets demonstrate that MorpMamba achieves superior parametric efficiency compared to traditional CNN and transformer models while maintaining high accuracy. The code will be made publicly available at https://***/mahmad000/MorpMamba. © 2024, CC BY.

关键词： Hyperspectral imaging

来源：评论

学校读者我要写书评

暂无评论

Chinese News Text Classification and Its Application Based on Combined-Convolutional Neural Network

引用

Journal of computers (Taiwan) 2022年第4期33卷 1-14页

作者： Liu, Kai-Feng Zhang, Yu Zhang, Quan-Xin Wang, Yan-Ge Gao, Kai-Long Jiangsu Vocational College of Finance and Economics Huaian223003 China School of Electrical and Information Engineering Beijing Key Laboratory of Intelligent Processing for Building Big Data Beijing University of Civil Engineering and Architecture Beijing100044 China State Key Laboratory in China for GeoMechanics and Deep Underground Engineering China University of Mining & Technology Beijing100083 China School of Computer Science and Technology Beijing Institute of Technology Beijing100081 China

A method based on combined-convolutional neural network (Combined-CNN) for Chinese news text classification is proposed. First of all, in order to solve the problem of a lack of special term set for Chinese news classification, a vocabulary suitable for Chinese long text classification is made by constructing a data index method. The Word2Vec pre-trained model was used to embed the text features word vectors. Second, by optimizing the structure of the classical convolutional neural network (CNN) model, a new idea of Combined-CNN model is proposed, which solves the problem of incomplete feature extraction of local text blocks and improves the accuracy rate of Chinese news text classification. Effective model regularization and RAdam optimization algorithm are designed in the model to enhance the model training effect. The experimental results show that the precision of the Combined-CNN model for Chinese news text classification reaches 93.69%. Compared with traditional machine learning methods and deep learning algorithms, the accuracy rate is improved by a maximum of 11.82% and 1.9%, respectively, and it is better than the comparison model in Recall and F-Measure. Finally, the Chinese news classification algorithm of the Combined-CNN is applied to realize a personalized recommendation system. © 2022 computer Society of the Republic of China. All rights reserved.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

SynFacePAD 2023: Competition on Face Presentation Attack Detection Based on Privacy-aware Synthetic Training Data

arXiv

引用

arXiv 2023年

作者： Fang, Meiling Huber, Marco Fierrez, Julian Ramachandra, Raghavendra Damer, Naser Alkhaddour, Alhasan Kasantcev, Maksim Pryadchenko, Vasiliy Yang, Ziyuan Huangfu, Huijie Chen, Yingyu Zhang, Yi Pan, Yuchen Jiang, Junjun Liu, Xianming Sun, Xianyun Wang, Caiyong Liu, Xingyu Chang, Zhaohua Zhao, Guangzhe Tapia, Juan Gonzalez-Soler, Lazaro Aravena, Carlos Schulz, Daniel Fraunhofer Institute for Computer Graphics Research IGD Darmstadt Germany Department of Computer Science TU Darmstadt Darmstadt Germany Biometrics and Data Pattern Analytics Lab Universidad Autonoma de Madrid Spain Norway ID R&D Inc. New York United States School of Cyber Science and Engineering Sichuan University Chengdu China School of Computer Science and Technology Harbin Institute of Technology Harbin China School of Electrical and Information Engineering Beijing University of Civil Engineering and Architecture China Biometrics and Security Research Group Hochschule Darmstadt Darmstadt Germany I+D Vision Center Santiago Chile

This paper presents a summary of the Competition on Face Presentation Attack Detection Based on Privacy-aware Synthetic Training Data (SynFacePAD 2023) held at the 2023 International Joint Conference on Biometrics (IJCB 2023). The competition attracted a total of 8 participating teams with valid submissions from academia and industry. The competition aimed to motivate and attract solutions that target detecting face presentation attacks while considering synthetic-based training data motivated by privacy, legal and ethical concerns associated with personal data. To achieve that, the training data used by the participants was limited to synthetic data provided by the organizers. The submitted solutions presented innovations and novel approaches that led to outperforming the considered baseline in the investigated benchmarks. © 2023, CC BY-NC-SA.

关键词： Competition

来源：评论

学校读者我要写书评

暂无评论

SynFacePAD 2023: Competition on Face Presentation Attack Detection Based on Privacy-aware Synthetic Training Data

SynFacePAD 2023: Competition on Face Presentation Attack Det...

引用

IEEE International Joint Conference on Biometrics (IJCB)

作者： Meiling Fang Marco Huber Julian Fierrez Raghavendra Ramachandra Naser Damer Alhasan Alkhaddour Maksim Kasantcev Vasiliy Pryadchenko Ziyuan Yang Huijie Huangfu Yingyu Chen Yi Zhang Yuchen Pan Junjun Jiang Xianming Liu Xianyun Sun Caiyong Wang Xingyu Liu Zhaohua Chang Guangzhe Zhao Juan Tapia Lazaro Gonzalez-Soler Carlos Aravena Daniel Schulz Fraunhofer Institute for Computer Graphics Research IGD Darmstadt Germany Department of Computer Science TU Darmstadt Darmstadt Germany Biometrics and Data Pattern Analytics Lab Universidad Autonoma de Madrid Spain Norwegian University of Science and Technology (NTNU) Norway ID R&D Inc New York US School of Cyber Science and Engineering Sichuan University Chengdu China School of Computer Science and Technology Harbin Institute of Technology Harbin China School of Electrical and Information Engineering Beijing University of Civil Engineering and Architecture China Biometrics and Security Research Group Hochschule Darmstadt Darmstadt Germany I+D Vision Center Santiago Chile

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：