检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Liu, Baokai He, Fengjie Du, Shiqiang Li, Jiacheng Liu, Wenjie Key Laboratory of China’s Ethnic Languages and Information Technology of Ministry of Education Chinese National Information Technology Research Institute Northwest Minzu University Gansu Lanzhou730030 China China Mobile Design Institute Co. Ltd. Shaanxi Branch Shaanxi Xi’an710000 China College of Mathematics and Computer Science Northwest Minzu University Gansu Lanzhou730030 China

Small object detection has important application value in the fields of autonomous driving and drone scene analysis. As one of the most advanced object detection algorithms, YOLOv3 suffers some challenges when detecting small objects, such as the problem of detection failure of small objects and occluded objects. To solve these problems, an improved YOLOv3 algorithm for small object detection is proposed. In the proposed method, the dilated convolutions mish (DCM) module is introduced into the backbone network of YOLOv3 to improve the feature expression ability by fusing the feature maps of different receptive fields. In the neck network of YOLOv3, the convolutional block attention module (CBAM) and multi-level fusion module are introduced to select the important information for small object detection in the shallow network, suppress the uncritical information, and use the fusion module to fuse the feature maps of different scales, so as to improve the detection accuracy of the algorithm. In addition, the Soft-NMS and Complete-IoU (CloU) strategies are applied to candidate frame screening, which improves the accuracy of the algorithm for the detection of occluded objects. The ablation experiment of the MS COCO2017 object detection task proves the effectiveness of several modules introduced in this paper for small object detection. The experimental results on the MS COCO2017, VOC2007, and VOC2012 datasets show that the Average Precision (AP) of this method is 16.5%, 8.71%, and 9.68% higher than that of YOLOv3, respectively. © 2022, CC BY.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Mechanical memory of excitable cells induced by mechanosensitive channel inactivation

引用

Biophysical Journal 2025年

作者： Jinjiang Xie Fangtao Mao Yuehua Yang Hongyuan Jiang CAS Key Laboratory of Mechanical Behavior and Design of Materials CAS Center for Excellence in Complex System Mechanics Department of Modern MechanicsUniversity of Science and Technology of China Hefei Anhui 230026 China

Recent experiments have unveiled a fascinating phenomenon: excitable cells exhibit mechanical memory, whereby their present excitation behaviours strongly depend on their past mechanical experiences. However, the underlying mechanism of this phenomenon remains elusive. Here, we introduce an electromechanical framework that integrates mechanical cell deformation, state transformations of mechanosensitive (MS) channels (such as Piezo channels), and transmembrane ion fluxes. We reveal that MS channel inactivation yields a history-dependent excitation dynamics, characterized by a progressive decline in subsequent activated currents with increasing amplitude, speed, and duration of prior mechanical stimuli. Moreover, MS channel inactivation in preceding stimulation results in a refractory period during which cells cannot elicit new action potentials upon subsequent mechanical stimuli. Finally, we show that cells can adapt to preceding mechanical stimulation due to inactivation of MS channels, resulting in a higher activated threshold stimulation. Thus, MS channel inactivation favors the reduction of firing activities in response to prolonged and repeated mechanical stimuli (“neural adaptation”), which may protect neurons against over-activation and damage. We then conduct two virtual experiments to predict how changes in mechanical properties of neurons modify their excitation behaviours. These findings together emphasize a critical role of MS channel inactivation in governing the mechanical memory and neural adaptation of excitable cells, shedding new light on the intricate interplay between mechanical forces and cellular electrical responses.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dunhuang murals contour generation network based on convolution and self-attention fusion

arXiv

引用

arXiv 2022年

作者： Liu, Baokai He, Fengjie Du, Shiqiang Zhang, Kaiwu Wang, Jianhua Key Laboratory of China’s Ethnic Languages and Information Technology of Ministry of Education Chinese National Information Technology Research Institute Northwest Minzu University Gansu Lanzhou730030 China China Mobile Design Institute Co. Ltd. Shaanxi Branch Shaanxi Xi’an710000 China College of Mathematics and Computer Science Northwest Minzu University Gansu Lanzhou730030 China

Dunhuang murals are a collection of Chinese style and national style, forming an autonomous Chinese-style Buddhist art. It has a very great historical and cultural value and an important research value. Among them, the lines of Dunhuang murals are extremely general and expressive. It reflects the character’s distinct character and intricate inner emotions. Hence, the outline drawing of the murals is of great importance in search of the Dunhuang culture. The generation of contours of Dunhuang murals belongs to the detection of image contours, which is an important branch of computer vision, aimed at extract salient contour information from the images. Although convolution-based deep learning networks have achieved good results in extracting image contours by exploring the contextual and semantic features of images. However, as the receptive field broadens, some detailed local information is lost. As a result, it is impossible for them to generate reasonable outline drawings of murals. In this paper, we propose a novel edge detector based on self-attention combined with convolution to generate line drawings of Dunhuang murals. Compared with existing edge detection methods, firstly, a new residual self-attention and convolution mixed module(Ramix) is proposed to merge local and global features into feature maps. Secondly, a novel densely connected backbone extraction network is designed to efficiently propagate information rich in edge features from shallow layers into deep layers. It is apparent from the experimental results that the method of this paper can generate richer and sharper edge maps on several standard edge detection datasets. In addition, tests on the Dunhuang mural dataset show that our method can achieve very competitive performance. © 2022, CC BY.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Joint Compression and Deadline Optimization for Wireless Federated Learning

arXiv

引用

arXiv 2023年

作者： Zhang, Maojun Li, Yang Liu, Dongzhu Jin, Richeng Zhu, Guangxu Zhong, Caijun Quek, Tony Q.S. The College of information Science and Electronic Engineering Zhejiang University Hangzhou China China Academy of Information and Communications Technology Beijing China School of Computing Science University of Glasgow United Kingdom The Department of Information and Communication Engineering Zhejiang University Hangzhou310007 China The Zhejiang–Singapore Innovation and AI Joint Research Lab Hangzhou310007 China Hangzhou310007 China The Shenzhen Research Institute of Big Data Guangdong Shenzhen518172 China The Peng Cheng Laboratory Guangdong Shenzhen518055 China Guangdong Guangzhou510555 China The Singapore University of Technology and Design Singapore487372 Singapore

Federated edge learning (FEEL) is a popular distributed learning framework for privacy-preserving at the edge, in which densely distributed edge devices periodically exchange model-updates with the server to complete the global model training. Due to limited bandwidth and uncertain wireless environment, FEEL may impose heavy burden to the current communication system. In addition, under the common FEEL framework, the server needs to wait for the slowest device to complete the update uploading before starting the aggregation process, leading to the straggler issue that causes prolonged communication time. In this paper, we propose to accelerate FEEL from two aspects: i.e., 1) performing data compression on the edge devices and 2) setting a deadline on the edge server to exclude the straggler devices. However, undesired gradient compression errors and transmission outage are introduced by the aforementioned operations respectively, affecting the convergence of FEEL as well. In view of these practical issues, we formulate a training time minimization problem, with the compression ratio and deadline to be optimized. To this end, an asymptotically unbiased aggregation scheme is first proposed to ensure zero optimality gap after convergence, and the impact of compression error and transmission outage on the overall training time are quantified through convergence analysis. Then, the formulated problem is solved in an alternating manner, based on which, the novel joint compression and deadline optimization (JCDO) algorithm is derived. Numerical experiments for different use cases in FEEL including image classification and autonomous driving show that the proposed method is nearly 30X faster than the vanilla FedSGD algorithm, and outperforms the state-of-the-art schemes. © 2023, CC BY.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Dynamic responses of 2-D fractional medium subjected to impact

引用

International Journal of Mechanical sciences 2025年 300卷

作者： Yuan, Liangzhu Xu, Songlin Yang, Haifeng Chen, Meiduo Lu, Jianhua Xie, Yushan Xiong, Ying Wang, Pengfei CAS key Laboratory for Mechanical Behavior and Design of Materials Department of Modern Mechanics University of Science and Technology of China Anhui Hefei230027 China China Earthquake Administration Beijing100036 China

It is of great significance to establish a new approach to investigate the 2-D dynamic responses of two-dimensional metamaterials with mesoscopic discontinuous structure due to their excellent anti-impact performance. The spatial fractional governing equations for the 2-D orthotropic medium subjected to impact are deduced, and their finite difference forms are given accordingly. The dynamic responses of the 2-D fractional medium are related to two fractional parameters, i.e., the fractional orders (α1 and α2) and the characteristic length (Δl1 and Δl2). The 2-D fractional medium shows greater flexibility in response simulation than the 1-D fractional medium. The plane-wave velocities of the 2-D fractional medium are obtained from the characteristics line method, and agree well with the numerical results. The oyster shell samples are impacted by the CO2 pulse laser, and their 2-D dynamic responses are measured by the two-point VISAR system. As the sample density increases, the velocity amplitudes decrease, and the amplitudes of the two signals become closer. The nacre-rich samples with higher density show obvious orthotropic properties, which is more suitable for the 2-D orthotropic fractional model, while the 2-D isotropic fractional model is more suitable for the chalk-rich samples with lower density. The 2-D orthotropic fractional model shows greater flexibility in fitting dynamic responses. The statistical relations of the fractional orders with the fractal dimension of the oyster shell samples are obtained by the FEM and by the fractional models. The fractional orders evaluated by the 2-D fractional model are much better than those evaluated by the 1-D fractional model. It provides a new approach to understanding the dynamic responses of 2-D discontinuous medium. © 2025 Elsevier Ltd

关键词： Fractal dimension

来源：评论

学校读者我要写书评

暂无评论

Optimal Status Updates for Minimizing Age of Correlated Information in IoT Networks with Energy Harvesting Sensors

arXiv

引用

arXiv 2023年

作者： Xu, Chao Zhang, Xinyan Yang, Howard H. Wang, Xijun Pappas, Nikolaos Niyato, Dusit Quek, Tony Q.S. The School of Information Engineering Northwest A&F University Shaanxi China The Key Laboratory of Agricultural Internet of Things Ministry of Agriculture and Rural Affairs Northwest A&F University Yangling712100 China The Zhejiang University University of Illinois Urbana-Champaign Institute Zhejiang University Haining China The School of Electronics and Information Technology SYSU Guangzhou China The Department of Computer and Information Science Linköping University Linköping Sweden The School of Computer Science and Engineering Nanyang Technological University Singapore The Singapore University of Technology and Design Singapore487372 Singapore The Yonsei Frontier Lab Yonsei University Korea Republic of

Many real-time applications of the Internet of Things (IoT) need to deal with correlated information generated by multiple sensors. The design of efficient status update strategies that minimize the Age of Correlated Information (AoCI) is a key factor. In this paper, we consider an IoT network consisting of sensors equipped with the energy harvesting (EH) capability. We optimize the average AoCI at the data fusion center (DFC) by appropriately managing the energy harvested by sensors, whose true battery states are unobservable during the decision-making process. Particularly, we first formulate the dynamic status update procedure as a partially observable Markov decision process (POMDP), where the environmental dynamics are unknown to the DFC. In order to address the challenges arising from the causality of energy usage, unknown environmental dynamics, unobservability of sensors’ true battery states, and large-scale discrete action space, we devise a deep reinforcement learning (DRL)-based dynamic status update algorithm. The algorithm leverages the advantages of the soft actor-critic and long short-term memory techniques. Meanwhile, it incorporates our proposed action decomposition and mapping mechanism. Extensive simulations are conducted to validate the effectiveness of our proposed algorithm by comparing it with available DRL algorithms for POMDPs. Copyright © 2023, The Authors. All rights reserved.

关键词： Energy harvesting

来源：评论

学校读者我要写书评

暂无评论

AI-based Automatic Segmentation of Prostate on Multi-modality Images: A Review

arXiv

引用

arXiv 2024年

作者： Jin, Rui Li, Derun Xiang, Dehui Zhang, Lei Zhou, Hailing Shi, Fei Zhu, Weifang Cai, Jing Peng, Tao Chen, Xinjian School of Future Science and Engineering Soochow University Suzhou China Department of Urology Peking University First Hospital Xicheng District Beijing100034 China MIPAV Lab The School of Electronic and Information Engineering Soochow University Jiangsu China Graduate Program of Medical Physics and Data Science Research Center Duke Kunshan University Jiangsu Kunshan215316 China Department of Mechanical Engineering and Product Design Engineering Swinburne University of Technology Melbourne Australia Department of Health Technology and Informatics The Hong Kong Polytechnic University Hong Kong Department of Radiation Oncology UT Southwestern Medical Center DallasTX United States The State Key Laboratory of Radiation Medicine and Protection Soochow University Jiangsu China

Prostate cancer represents a major threat to health. Early detection is vital in reducing the mortality rate among prostate cancer patients. One approach involves using multi-modality (CT, MRI, US, etc.) computer-aided diagnosis (CAD) systems for the prostate region. However, prostate segmentation is challenging due to imperfections in the images and the prostate’s complex tissue structure. The advent of precision medicine and a significant increase in clinical capacity have spurred the need for various data-driven tasks in the field of medical imaging. Recently, numerous machine learning and data mining tools have been integrated into various medical areas, including image segmentation. This article proposes a new classification method that differentiates supervision types, either in number or kind, during the training phase. Subsequently, we conducted a survey on artificial intelligence (AI)-based automatic prostate segmentation methods, examining the advantages and limitations of each. Additionally, we introduce variants of evaluation metrics for the verification and performance assessment of the segmentation method and summarize the current challenges. Finally, future research directions and development trends are discussed, reflecting the outcomes of our literature survey, suggesting high-precision detection and treatment of prostate cancer as a promising avenue. Impact Statement — Prostate cancer remains a significant health threat, necessitating early detection to reduce mortality rates. Current diagnostic approaches, particularly in medical imaging, face challenges in accurate prostate segmentation due to the organ's complex structure and imaging imperfections. Our review paper delves into AI-based automatic segmentation methods across various imaging modalities (CT, MRI, US), emphasizing their strengths, weaknesses, and the specific challenges they address. By classifying segmentation techniques based on supervision levels, we provide a comprehensive analys

关键词： Computer aided diagnosis

来源：评论

学校读者我要写书评

暂无评论

Optimising Urban Energy Dynamics: A Multifactorial Approach to Building Energy Consumption and Pv Generation in Urban Blocks

SSRN

引用

SSRN 2024年

作者： Li, Gaomei Zhou, Huangwanjin Kang, Jian Sun, Hao Xu, XiaoDong Chen, Yixing Zhong, Hua Xu, Shen School of Architecture & Urban Planning Huazhong University of Science and Technology Wuhan430074 China State Key Laboratory of Subtropical Building and Urban Science Guangzhou510641 China Hubei Engineering and Technology Research Center of Urbanization Wuhan430074 China LondonWC1E 6BT United Kingdom School of Architecture Design and Built Environment Nottingham Trent University NottinghamNG1 4FQ United Kingdom Department of Architecture and Built Environment University of Nottingham NottinghamNG7 2RD United Kingdom College of Civil Engineering Hunan University Changsha410082 China School of Architecture Southeast University Nanjing210096 China

Urban block morphology and function are key factors affecting building energy consumption and photovoltaic (PV) generation potential. This study investigates the synergistic effect of urban block morphology and function on building energy consumption and PV power generation potential. The research employs a k-means clustering algorithm to categorize 150 urban blocks in Wuhan, China, and proposes a workflow for evaluating energy use, solar potential, and load balancing on a block scale. The findings reveal that office blocks consume 148.90% more energy annually than residential blocks, while commercial blocks consume 323.06% more. The block with the highest PV energy generation potential was multi storey slab commercial streets, which was 422.06% higher than high-rise pavilion A office block. The study also found that multi-storey slab and multi-storey courtyard residential blocks could achieve a balance between energy supply and demand with over 100% PV substitution rates while the high-rise pavilion A office block had only 14.63%. The study contributes a scalable model for assessing urban energy dynamics considering block morphology and function over time. The findings can guide sustainable urban planning to balance energy supply and demand. © 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

Effect of the atomic mass mismatch in nanopillars on surface localization

Thermal Advances

引用

Thermal Advances 2024年 1卷

作者： Zhuo Zhao Haobo Fan Yiyi Li Meng Yuan Daxin Liang Jian Zhang Key Laboratory of Bio-Based Material Science & Technology Ministry of Education Northeast Forestry University Harbin 150040 China Yangtze Delta Region Academy of Beijing Institute of Technology (Jiaxing) Jiaxing 314019 China School of Astronautics Harbin Institute of Technology Harbin 150001 China ChinaState Key Laboratory of Advanced Nuclear Energy Technology Nuclear Power Institute of China Chengdu 610213 China CNNC Key Laboratory of Nuclear Reactor Thermal and Hydraulic Engineering Nuclear Power Institute of China Chengdu 610213 China

In recent years, nanophononic metamaterials (NPMs) have attracted great interest from researchers. Theoretical calculations and experimental studies have revealed the mechanism of reducing thermal conductivity through localized resonant hybridization. In addition, researchers have studied the effects of nanopillar height, spacing, and atomic mass in nanopillars on surface localization. The atomic mass mismatch in nanopillars has important application in practice. Taking the silicon-based nanopillar structure as an example, this paper studies the effects of atomic mass mismatch in nanopillars on surface localization effect and heat flux. The results show that atomic mass mismatch in nanopillars can enhance the surface localization effect, and the greater the difference in atomic mass in nanopillars, the stronger the surface localization. Finally, the effect of atomic mass mismatch in nanopillars on heat flux was calculated by nonequilibrium molecular dynamics. The results of this study can provide useful guidance for the design of nanoscale heat flux regulation devices.

关键词： Crystalline silicon Phononic metamaterials Atomic mass mismatch Phonon localization Molecular dynamics

来源：评论

学校读者我要写书评

暂无评论

Interfacial magnetic spin Hall effect in van der Waals Fe3GeTe2/MoTe2 heterostructure

arXiv

引用

arXiv 2024年

作者： Dai, Yudi Xiong, Junlin Ge, Yanfeng Cheng, Bin Wang, Lizheng Wang, Pengfei Liu, Zenglin Yan, Shengnan Zhang, Cuiwei Xu, Xianghan Shi, Youguo Cheong, Sang-Wook Xiao, Cong Yang, Shengyuan A. Liang, Shi-Jun Miao, Feng National Laboratory of Solid State Microstructures Institute of Brain-Inspired Intelligence School of Physics Collaborative Innovation Center of Advanced Microstructures Nanjing University Nanjing210093 China Research Laboratory for Quantum Materials Singapore University of Technology and Design Singapore Singapore Institute of Interdisciplinary Physical Sciences School of Science Nanjing University of Science and Technology Nanjing210094 China Institute of Physics Chinese Academy of Sciences Beijing100190 China Center for Quantum Materials Synthesis Department of Physics and Astronomy Rutgers The State University of New Jersey PiscatawayNJ08854 United States Institute of Applied Physics and Materials Engineering University of Macau Taipa Macau China Department of Physics University of Hong Kong Hong Kong Hong Kong HKU-UCAS Joint Institute of Theoretical and Computational Physics at Hong Kong Hong Kong Hong Kong

The spin Hall effect (SHE) allows efficient generation of spin polarization or spin current through charge current and plays a crucial role in the development of spintronics. While SHE typically occurs in non-magnetic materials and is time-reversal even, exploring time-reversal-odd (T-odd) SHE, which couples SHE to magnetization in ferromagnetic materials, offers a new charge-spin conversion mechanism with new functionalities. Here, we report the observation of giant Todd SHE in Fe3GeTe2/MoTe2 van der Waals heterostructure, representing a previously unidentified interfacial magnetic spin Hall effect (interfacial-MSHE). Through rigorous symmetry analysis and theoretical calculations, we attribute the interfacial-MSHE to a symmetry-breaking induced spin current dipole at the vdW interface. Furthermore, we show that this linear effect can be used for implementing multiply-accumulate operations and binary convolutional neural networks with cascaded multi-terminal devices. Our findings uncover an interfacial T-odd charge-spin conversion mechanism with promising potential for energy-efficient in-memory computing. © 2024, CC BY-NC-SA.

关键词： Spin Hall effect

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：