检索结果-内蒙古大学图书馆

An Accurate Tunnel Crack Identification Method Integrating Local Segmentation and Global Fusion Detection

IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES 2025年第4期E108A卷 606-612页

作者： Wang, Baoxian Gao, Ze Xu, Hongbin Qin, Shoupeng Tan, Zhao Shi, Xuchao Shijiazhuang Tiedao Univ Key Lab Struct Hlth Monitoring & Control Shijiazhuang 050043 Peoples R China Shenzhen Univ Shenzhen 518060 Peoples R China China Railway Design Corp Tianjin 300251 Peoples R China Hebei Jinbo Elevator Intelligent Equipment Co Ltd Shijiazhuang 050043 Peoples R China

This paper proposes a novel approach for tunnel crack identification, employing local segmentation and global fusion detection. Initially, a local segmentation network is constructed using weights from an encoding layer pre-trained on numerous non-tunnel cracks, with only a limited number of tunnel crack samples. The input image is divided, and the local segmentation network performs pixel segmentation on these sub-images, with the sub-results stitched together to ensure accurate identification of all suspicious crack pixels. Subsequently, a global fusion detector is introduced, comprising two sub-models: Sub-model 1 extracts total crack targets within the stitched results, while Sub-model 2 detects possible false alarms from regular-shaped areas. The results from both sub-models are combined to effectively reduce the false alarm rate and ensure accurate segmentation results of cracks. Experimental findings on actual tunnel images demonstrate that the "segmentation before localization" method proposed in this paper achieves superior recognition accuracy and IOU ratio compared to the Unet3+, DeeplabV3+, and "localization before segmentation" Mask-RCNN algorithms. Specifically, the proposed method yields an accuracy improvement of 3.81% over the Unet3+ network, 2.71% over the DeeplabV3++ network, and 1.93% over the Mask-RCNN network. Moreover, noise interference from bolt repair areas is effectively mitigated, enhancing the method's engineering applicability.

关键词： crack identification local segmentation global fusion detection transfer learning

来源：评论

学校读者我要写书评

暂无评论

Multimodal fusion of deep neural networks for transmission line flashover classification

引用

INTERNATIONAL JOURNAL OF PARALLEL EMERGENT AND DISTRIBUTED SYSTEMS 2025年第3期40卷 222-237页

作者： Feng, Bo Zhang, Wei Xia, Xiaofei Tang, Jie Liu, Peng Guangxi Power Grid Co Ltd Guangxi Key Lab Intelligent Control & Maintenance Elect Power Res Inst Nanning Peoples R China Guangxi Power Grid Co Ltd Guangxi Power Grid Equipment Monitoring & Diag Eng Elect Power Res Inst Nanning Peoples R China Guangxi Power Grid Co Ltd Nanning Peoples R China Guangxi Power Grid Co Ltd Guangxi Key Lab Intelligent Control & Maintenance Elect Power Res Inst Nanning 530023 Guangxi Peoples R China

The power system plays a vital role in modern society. However, the occurrence of transmission line flashover faults seriously threatens the safety and stability of the power system. In this paper, a deep neural network flashover classification model based on multimodal attention aggregation is proposed. By using the model, the problems of insufficient feature extraction capability of the unimodal flashover data model and easy loss of key feature information in the process of audiovisual multimodal flashover data alignment are solved. Experiments on the arc- flashover fault dataset show that the flashover fault detection model proposed in this paper can be used stably and effectively for flashover fault detection. {GRAPHICAL ABSTRACT}

关键词： Flashover fault multimodal data fusion cross-modal attention multi-attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Incentive-Driven Partial Offloading and Resource Allocation in Vehicular Edge Computing Networks

引用

IEEE INTERNET OF THINGS JOURNAL 2025年第8期12卷 11023-11035页

作者： Meng, Deng Guo, Jianmeng Zhou, Huan Zhang, Yao Zhao, Liang Shu, Yuanchao Fan, Xinggang China Three Gorges Univ Coll Econ & Management Yichang 443002 Peoples R China China Three Gorges Univ Hubei Key Lab Intelligent Vis Based Monitoring Hyd Yichang 443002 Peoples R China China Three Gorges Univ Hubei Key Lab Intelligent Vis Based Monitoring Hyd Yichang 443002 Peoples R China China Three Gorges Univ Coll Comp & Informat Technol Yichang 443002 Peoples R China Northwestern Polytech Univ Sch Comp Sci Xian 710129 Peoples R China Zhejiang Univ State Key Lab Ind Control Technol Hangzhou 310027 Peoples R China Zhejiang Univ Technol Coll Zhijiang Shaoxing 312030 Peoples R China

Vehicle edge computing can effectively ensure the quality of experience for user vehicles (UVs), but road side units (RSUs) with limited resources may not be able to handle intensive tasks under high traffic conditions. In this case, worker vehicles (WVs) with idle resources can share resources to alleviate the pressure on RSUs. However, selfish WVs may be reluctant to share idle computation resources without any rewards. In addition, the optimization problems in previous research are relatively simple and cannot be applied to complex scenarios. To address the above challenges, we propose an incentive-driven partial offloading framework aiming to maximize social welfare. In particular, the computing service provider (CSP) managing RSUs first determines resource prices and offloading rates with UVs, while also determining contract terms with WVs. Then, it generates the optimal task scheduling strategy and notifies the UVs to offload tasks to the corresponding WVs. Considering that maximizing social welfare is a mixed-integer nonlinear programming (MINLP) problem, we design the hybrid proximal policy optimization (HPPO)-based task offloading and resource allocation algorithm (HORA) with a hybrid action space to directly solve the original problem. Finally, extensive simulation results show that HORA outperforms other baseline methods across various scenarios, and the contract terms meet the constraints of individual rationality (IR) and incentive compatibility (IC).

关键词： Resource management Contracts Servers Costs Optimization Internet of Things Fans Energy consumption Edge computing Sun Contract theory hybrid proximal policy optimization (HPPO) partial offloading resource allocation social welfare

来源：评论

学校读者我要写书评

暂无评论

Coarray Tensor Train Aided Target Localization for Bistatic MIMO Radar

引用

IEEE SIGNAL PROCESSING LETTERS 2025年 32卷 46-50页

作者： Xie, Qianpeng Wen, Fangqing Xie, Xiaoxia Wang, Zhanling Pan, Xiaoyi Natl Univ Def Technol State Key Lab Complex Electromagnet Environm Effec Changsha 410073 Peoples R China Hubei Univ Automot Technol Inst Vehicle Informat Control & Network Technol Shiyan 442002 Peoples R China China Three Gorges Univ Hubei Key Lab Intelligent Vis Based Monitoring Hyd Yichang 443002 Peoples R China

In this letter, a coarray Tensor Train (TT) decomposition method is proposed to locate the targets in bistatic Multiple-Input Multiple-Output (MIMO) radar with uniform planar array (UPA) geometry. Initially, a five-dimensional (5-D) tensor model is established to preserve the multi-dimensional structure in the receive part. Subsequently, a four-dimensional (4-D) low-rank tensor can be obtained by removing redundant elements in the difference coarray of the eight-dimensional (8-D) covariance tensor. The TT-SVD algorithm is then applied to convert this 4-D tensor into a sequential product of lower-order TT-cores. Compared to CANDECOMP/PARAFAC decomposition (CPD) and Tucker decomposition (TD), the advantages of TTD include flexible multi-way data representation and mitigation of the curse of dimensionality. Furthermore, by using the relationship between Vandermonde factors matrices and TT-cores, a new method by combining different TT-cores is devised to estimate the angle parameters. Simulation results confirm the superiority of the proposed TT-aided method over other tensor-based methods for bistatic MIMO radar.

关键词： Tensors MIMO radar Transmission line matrix methods Matrix decomposition Vectors Planar arrays Estimation Covariance matrices Location awareness Geometry Bistatic MIMO radar coarray tensor angle parameter estimation tensor train decomposition (TTD) TT-SVD algorithm

来源：评论

学校读者我要写书评

暂无评论

Tensor decomposition estimator applied to coherent targets in EMVS-MIMO radar

引用

PHYSICAL COMMUNICATION 2025年 69卷

作者： Yu, Zhiyu Chen, Qiulin Wen, Fangqing Li, Zhengzhou Chongqing Univ Sch Microelect & Commun Engn Shazheng St 174 Chongqing 400044 Peoples R China Hubei Univ Automot Technol Inst Vehicle Informat Control & Network Technol 167 Checheng West Rd Shiyan 442002 Hubei Peoples R China China Three Gorges Univ Hubei Key Lab Intelligent Vis Based Monitoring Hyd 8 Daxue Rd Yichang 443002 Hubei Peoples R China

Electromagnetic vector sensor (EMVS) arrays in multiple-input multiple-output (MIMO) radar systems have gained substantial attention for their enhanced capabilities in target detection and localization over the past decade. However, conventional direction of arrival (DOA) estimation algorithms exhibit significant limitations when processing coherent targets. In response to these challenges, this paper introduces a novel algorithm leveraging tensor decomposition to enhance DOA estimation accuracy within monostatic EMVS-MIMO radar systems. Initially, we employ a specific sequence to rearrange the original array output, thereby linking the source matrix with the spatial response of the transmitting or receiving array. This rearrangement effectively addresses the rank deficiency issue in the covariance matrix. Subsequently, the restructured model undergoes parallel factor (PARAFAC) decomposition to yield high-precision estimates of the factor matrices. Ultimately, 2D-DOA estimation is accomplished by applying the normalized vector cross product (NVCP) technique to the polarization response factor matrices. In contrast to existing algorithms for coherent targets estimation, the proposed method ensures no information loss, achieves superior angle estimation accuracy, and is applicable to arbitrary sensor geometries. Numerical simulations substantiate the effectiveness and superiority of the proposed algorithm.

关键词： Electromagnetic vector sensor Direction of arrival estimation PARAFAC decomposition Coherent targets MIMO radar

来源：评论

学校读者我要写书评

暂无评论

Challenges, Progress, and Prospects of Ultra-Long Deep Tunnels in the Extremely Complex Environment of the Qinghai-Xizang Plateau

引用

ENGINEERING 2025年第1期44卷 162-183页

作者： Zhao, Yong Du, Yanliang Yan, Qixiang Tibet Railway Construct Corp Lhasa 850000 Peoples R China Shenzhen Univ Natl Key Lab Green & Long Life Rd Engn Extreme Env Shenzhen 518060 Peoples R China Shijiazhuang Tiedao Univ Key Lab Struct Hlth Monitoring & Control Shijiazhuang 050043 Peoples R China Southwest Jiaotong Univ State Key Lab Intelligent Geotech & Tunnelling Chengdu 610031 Peoples R China

With the implementation of significant national strategies and rapid socioeconomic development, many ultra-long deep tunnels are being constructed in the Qinghai-Xizang Plateau region. However, the extreme complexity and variability of the environment in this region pose significant challenges to the safe construction and long-term operation of the planned or under-construction ultra-long deep tunnels. To address these complex technical challenges, this paper provides a detailed analysis of the complex climate and geology features of the Qinghai-Xizang Plateau during tunnel construction. The climate characteristics of the Qinghai-Xizang Plateau include severe coldness, low oxygen, and unpredictable weather changes. The geological characteristics include complex stress distributions caused by the intense internal and external dynamic coupling of tectonic plates, widespread active tectonic structures, frequent high-intensity earthquakes, fractured rock masses, and numerous active fault zones. Based on the analysis, this paper elaborates on potential sources of major disasters resulting from the characteristics of ultra-long deep tunnel projects in the Qinghai-Xizang Plateau region. These potential disaster sources include the crossing of active fault zones, high geostress rockbursts, large deformation disasters, high-pressure water surges, geothermal hazards, inadequate long-distance ventilation and oxygen supply, and multi-hazard couplings. In response to these challenges, this paper systematically summarizes the latest research progress and technological achievements in the domestic and international literature, and proposes innovative ideas and future development prospects for disaster monitoring and early warning, mechanized intelligent construction, long-term safety services, and emergency security and rescue. These innovative measures are intended to address the challenges of tunnel disaster prevention and control in the complex environment of the Qinghai-Xizang

关键词： The Qinghai-Xizang Plateau Ultra-long deep tunnels Multi-hazards coupling Active prevention and control Mechanization Intelligence

来源：评论

学校读者我要写书评

暂无评论

The Effects of Three Bean Shell Biochars Under Different Pyrolysis Temperatures on the Adsorption of Cd and Pb in Aqueous Solutions

引用

WATER 2025年第7期17卷 918-918页

作者： Shen, Tao Xia, Hongyu Zhang, Heyi Guang, Song Hu, Wenwen Zhao, Wenrui Zhao, Kuan Xiao, Xin Zhang, Shiwen Xu, Aiai Anqing Normal Univ Key Lab Intelligent Qual Monitoring & Soil Fertil Coll Resources & Environm Key Lab Aqueous Environm Protect & Pollut Control Anqing 246133 Peoples R China Anhui Univ Sci & Technol Sch Earth & Environm Huainan 232001 Peoples R China Fujian Acad Agr Sci Inst Resources Environm & Soil Fertilizer Fujian Key Lab Plant Nutr & Fertilizer Fuzhou 350013 Peoples R China

Biochar is an eco-friendly material that influences heavy metals adsorption. Three kinds of bean shell biochar, namely, peanut, pea, and soybean biochar, were prepared at pyrolysis temperatures of 400 degrees C, 500 degrees C, and 600 degrees C. The characteristics of these biochars and the physicochemical properties of the biochars were analyzed. The capacities of the different biochars to adsorb Pb and Cd from aqueous solution were determined. With increasing pyrolysis temperature, the relative content of carbon in the biochar increased and that of hydrogen decreased, the porosity decreased, and the specific surface area increased;accordingly, the adsorption capacity of the biochar for Pb and Cd increased. The Pb adsorption capacity of the peanut shell biochar prepared at 400 degrees C was lower than that of the other shell biochars when the initial Pb concentration was at a low concentration. Three adsorption isotherm models were used to fit the adsorption processes of Pb or Cd on the three different biochars. The Freundlich curves better fit the adsorption capacity of biochar for Cd, and the Freundlich and Langmuir curves better fit the adsorption capacity of biochar for Pb. This work provides a scientific basis for the rational selection of bean shell biochar used in metal-contaminated water.

关键词： bean shell biochar representation heavy metals adsorption

来源：评论

学校读者我要写书评

暂无评论

An Effective Mesh Optimization Strategy for Oil Immersed Transformer Thermal Fluid Field Calculation 11th

An Effective Mesh Optimization Strategy for Oil Immersed Tra...

引用

11th Frontier Academic Forum of Electrical Engineering, FAFEE 2024

作者： Zhang, Lei Li, Rui Chen, Liangyuan Sun, Zhiyuan Electric Power Research Institute of Guangxi Power Grid Co. Ltd Guangxi Nanning530023 China Guangxi Key Laboratory of Intelligent Control and Maintenance of Power Equipment Guangxi Nanning530023 China Guangxi Power Grid Equipment Monitoring and Diagnosis Engineering Technology Research Center Guangxi Nanning530023 China

ISBN: (纸本)9789819788279

A transformer thermal fluid field mesh optimization generation method based on the calculation results difference feedback is proposed. The number of meshes obtained by the local mesh refinement optimization based on the difference distribution feedback of results is 1.972 million, which is reduced to 72.7% of the number of meshes obtained based on the global refinement method. A temperature rise test platform is built in which the windings have embedded optical fiber temperature measurement system, and based on the test results under rated condition, the absolute errors between the calculated hot spots of HV and LV windings and the test values are 0.81% and 0.40%, respectively, which verify the effectiveness of the method. The thermal fluid field calculation mesh optimization generation method can reduce the number of calculation meshes while ensuring the calculation accuracy, and can provide a reference for mesh optimization control in transformer thermal fluid field calculation. © Beijing Paike Culture Commu. Co., Ltd. 2025.

关键词： Mesh generation

来源：评论

学校读者我要写书评

暂无评论

Lightweight skeleton-based action recognition model based on global-local feature extraction and fusion

引用

INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS 2025年第3期16卷 1477-1488页

作者： Deng, Zhe Wang, Yulin Wei, Xing Yang, Fan Zhao, Chong Lu, Yang Hefei Univ Technol Sch Mech Engn Hefei 230009 Anhui Peoples R China Hefei Univ Technol Sch Comp & Informat Hefei 230601 Anhui Peoples R China Hefei Univ Technol Intelligent Mfg Technol Res Inst Hefei 230051 Anhui Peoples R China Minist Educ Engn Res Ctr Safety Crit Ind Measurement & Control Hefei 230009 Anhui Peoples R China Anhui Mine IOT & Secur Monitoring Technol Key Lab Hefei 230088 Anhui Peoples R China

Skeleton-based action recognition has become a research hotspot in the field of computer vision because of its lightweight and strong anti-interference. However, there are disadvantages such as single feature extraction, limited expression ability, and low recognition accuracy. To solve these problems, we propose a lightweight Skeleton-based action recognition model based on global-local feature extraction and fusion (GLF-GCN). GLF-GCN includes a Feature extraction of non-connected nodes Module (Global-GCN), a Feature extraction of adjacent nodes Module (Local-GCN), and a Dynamic Fusion module. More specifically, Global-GCN combines one-dimensional convolution and shift operations to capture spatio-temporal dependencies across global nodes, using shift operations as a replacement for spatio-temporal graph convolution to reduce computational complexity. Meanwhile, Local-GCN captures temporal and spatial local information from first-order neighboring nodes. On this basis, Dynamic Fusion integrates global information based on joint hierarchy and local information based on body parts to discern the varying dependency relationships among different body parts and joints, improving the model's ability to interpret different skeleton action sequences. The experimental results on single stream and multi-stream data show that the proposed model has higher accuracy, which attains the state-of-the-art performance.

关键词： Graph convolutional neural networks Local feature extraction Feature fusion Shift graph convolution

来源：评论

学校读者我要写书评

暂无评论

Temporal and Spatial Effects of Extreme Drought Events on Human Epidemics over Ancient China in 1784-1787 CE

引用

ENVIRONMENTAL HEALTH 2025年第1期24卷 1-16页

作者： Hang, Xiaoyi Sun, Zhaobin He, Juan Xin, Jingyi Zhang, Shuwen Zhao, Yuxin Tan, Zhen Han, Ling Hao, Yu Beijing Univ Chinese Med Sch Tradit Chinese Med Beijing 100029 Peoples R China China Meteorol Adm Chinese Acad Meteorol Sci CAMS State Key Lab Severe Weather LASW Beijing 100081 Peoples R China Nanjing Univ Chinese Med Sch Tradit Chinese Med Nanjing 210023 Peoples R China Beijing Municipal Ecol & Environm Monitoring Ctr Beijing 100048 Peoples R China China Meteorol Adm Dept Forecasting & Networking Beijing 100081 Peoples R China Natl Inst Communicable Dis Control & Prevent Chinese Ctr Dis Control & Prevent Natl Key Lab Intelligent Tracking & Forecasting In Beijing 102206 Peoples R China

Extreme drought events can impact human health, notably triggering epidemics that impose significant global health and economic burdens. Understanding these effects and developing response strategies is crucial. However, there is limited epidemiological evidence on how climate change influenced ancient epidemics before large-scale urbanization and frequent population movements. In this study, we utilized the Reconstructed East Asian Climate Historical Encoded Series (REACHES) climate database and the self-constructed ancient Chinese epidemics database to examine extreme drought events in ancient China from 1784-1787 CE. We analyzed factors like grain prices, population density, and socioeconomic conditions to explore the temporal and spatial mechanism and influence the degree of extreme drought events on epidemics outbreaks. The results show that there is a clear positive link between drought and the spread of epidemics, with a notable one-year lag effect of drought. Drought impacts epidemics directly and indirectly through locust plague, famine, crop failure, and social turmoil, with famine being the most crucial factor. Official disaster management can mitigate epidemics. This study intuitively shows the relationship between extreme drought events and epidemics in ancient China and offering insights into the climate change-epidemic relationship. Placing the conclusions of this paper in a broader context has global implications, providing historical experience for polycrisis and modern challenges.

关键词： Extreme drought events Human epidemics Historical climate restoration Socioeconomic factors Ancient China

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：