检索结果-内蒙古大学图书馆

U-net of joint spatial domains with multi-scale atrous convolution for rectal image segmentation

Multimedia Tools and Applications 2025年 1-17页

作者： Rao, Yunbo Gao, Li Zeng, Shaoning Shao, Tingting Sun, Jihong School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu611731 China Yangtze Delta Region Institute University of Electronic Science and Technology of China Huzhou313000 China Zhejiang University School of Medicine Zhejiang Hangzhou310058 China

Medical image segmentation is very important for the diagnosis of related diseases. To reduce the labeling work of related medical images, numerous models based on U-Net have been proposed to achieve automatic segmentation of target regions. However, most of these models are only trained in one coordinate system, ignoring the joint effects of different spatial coordinate systems. In addition, most of the encoding modules in these models do not pay attention to multi-scale spatial information. Our proposed solution for the aforementioned challenges involves using U-Net model with joint spatial domains and multi-scale encoding module, which enables us to segment rectal image better. The model includes a self-designed multi-layer dilated convolution encoding module named AIR (Atrous Inception Residual Block), to achieve a multi-scale content fusion. Besides this, it utilizes the center point and polar coordinates to realize attention mechanism and rotation invariance. Furthermore, retraining the output of polar coordinate network with Cartesian coordinate system realizes the translation invariance of segmentation. Compared with the commonly used medical segmentation models, the dice coefficient of our model is improved by about 2% on our in-house rectal dataset. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Optimization of multi-state generation problem based on spatial information network topology

引用

High-Confidence Computing 2023年第1期3卷 58-65页

作者： Peng Yang JiaYing Zhang Shijie Zhou Jinyu Zhou School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu 610054China School of Mathematical Sciences University of Electronic Science and Technology of ChinaChengdu 611731China

Spatial information network is a kind of satellite network with high speed node movement and fast dynamic topology *** the increasing number of low-orbit satellites,the research on the subnets topology and dynamic optimization of space information networks has become an important direction to study the destructibility of spatial information *** this paper,two common objective functions in inter-satellite link assignment,network observation position and network communication factor are studied,and a multi-objective optimization model is *** first search,simulated annealing,NSGA-II and adaptive optimization simulated annealing were used to analyze and solve the *** comparing the solving efficiency of the model through simulation experiments,the difference of the results caused by the four algorithms is verified.

关键词： Spatial information network Simulated annealing NSGA-II

来源：评论

学校读者我要写书评

暂无评论

Imaging through scattering layers using a near-infrared low-spatial-coherence fiber random laser

引用

Chinese Physics B 2024年第10期33卷 226-232页

作者： Anda Shi Zeyu Wang Chenxi Duan Zhao Wang Weili Zhang School of Information and Communication Engineering University of Electronic Science&Technology of ChinaChengdu 611731China Yingcai Honors College University of Electronic Science&Technology of ChinaChengdu 611731China

Optical memory effect-based speckle-correlated technology has been developed for reconstructing hidden objectsfrom disordered speckle patterns,achieving imaging through scattering ***,the lighting efficiency and fieldof view of existing speckle-correlated imaging systems are ***,a near-infrared low spatial coherence fiberrandom laser illumination method is proposed to address the above *** the utilization of random Rayleighscattering within dispersion-shifted fibers to provide feedback,coupled with stimulated Raman scattering for amplification,a near-infrared fiber random laser exhibiting a high spectral density and extremely low spatial coherence is *** on the designed fiber random laser,speckle-correlated imaging through scattering layers is achieved,with highlighting efficiency and a large imaging field of *** work improves the performance of speckle-correlated imagingand enriches the research on imaging through scattering medium.

关键词： fiber random laser speckle-correlated imaging scattering medium spatial coherence

来源：评论

学校读者我要写书评

暂无评论

Domain-adaptive Solar Cell Surface Defect Detection Algorithm

Journal of Network Intelligence

引用

Journal of Network Intelligence 2024年第4期9卷 2588-2604页

作者： Zhang, Xiong Li, Ran-Ran Hou, Ting Shangguan, Hong Wu, Xiao-Jia Ning, Ai-Ping Wang, An-Hong School of electronic information engineering Taiyuan University of Science and Technology Taiyuan030024 China

Domain-adaptive networks have demonstrated outstanding performance in bridging interdomain disparities, emerging as a burgeoning research direction in the field of object detection in recent years. However, the substantial disparities between the distribution of solar cell defect detection data in real-world scenarios and that of training datasets significantly impact the accuracy and robustness of detection networks. To address this issue, this paper first constructs three sets of datasets, including "source domain-exposure domain dataset," " source domain-night domain dataset," and "source domain dataset-noise domain dataset." Secondly, an uncertainty-aware module is introduced into the feature extraction module of the baseline domain-adaptive multi-level entropy attention alignment (MEAA) model to enhance the network’s efficiency in utilizing features during interdomain image-level alignment. Lastly, an instance-level efficient region proposal network (RPN) alignment module is designed to improve the network’s instance-level prediction accuracy. Experimental results demonstrate that the proposed method significantly enhances domain adaptive detection accuracy across different domains and exhibits notable superiority compared to state-of-the-art domain adaptive detection networks. © 2024, Taiwan Ubiquitous information CO LTD. All rights reserved.

关键词： Defect detection Domain-adaptive networks Imagelevel alignment Instance-level alignment Interdomain differences

来源：评论

学校读者我要写书评

暂无评论

Improved Double Deep Q Network-Based Task Scheduling Algorithm in Edge Computing for Makespan Optimization

引用

Tsinghua science and technology 2024年第3期29卷 806-817页

作者： Lei Zeng Qi Liu Shigen Shen Xiaodong Liu School of Computer Science Nanjing University of Information Science and TechnologyNanjing 210044China School of Software Nanjing University of Information Science and TechnologyNanjing 210044China School of Information Engineering Huzhou UniversityHuzhou 313000China School of Computing Edinburgh Napier UniversityEdinburghEH105DTUK

Edge computing nodes undertake an increasing number of tasks with the rise of business ***,how to efficiently allocate large-scale and dynamic workloads to edge computing resources has become a critical *** study proposes an edge task scheduling approach based on an improved Double Deep Q Network(DQN),which is adopted to separate the calculations of target Q values and the selection of the action in two networks.A new reward function is designed,and a control unit is added to the experience replay unit of the *** management of experience data are also modified to fully utilize its value and improve learning *** learning agents usually learn from an ignorant state,which is *** such,this study proposes a novel particle swarm optimization algorithm with an improved fitness function,which can generate optimal solutions for task *** optimized solutions are provided for the agent to pre-train network parameters to obtain a better cognition *** proposed algorithm is compared with six other methods in simulation *** show that the proposed algorithm outperforms other benchmark methods regarding makespan.

关键词： edge computing task scheduling reinforcement learning makespan Double Deep Q Network(DQN)

来源：评论

学校读者我要写书评

暂无评论

Research on temperature distribution characteristics of oilimmersed power transformers based on fluid network decoupling

引用

High Voltage 2024年第5期9卷 1136-1148页

作者： Yongming Xu Ziyi Xu Congrui Ren Yaodong Wang School of Electrical and Information Engineering Changzhou Institute of TechnologyChangzhouChina School of Electrical and Electronic Engineering Harbin University of Science and TechnologyHarbinChina Department of Engineering Durham UniversityDurhamUK

Due to the complex structure and large size of large-capacity oil-immersed power transformers,it is difficult to predict the winding temperature distribution directly by numerical analysis.A 180 MVA,220 kV oil-immersed self-cooling power transformer is used as the research *** authors decouple the internal fluid domain of the power transformer into four regions:high voltage windings,medium voltage windings,low voltage windings,and radiators through fluid networks and establish the 3D fluidtemperature field numerical analysis model of the four regions,*** results of the fluid network model are used as the inlet boundary conditions for the 3D fluidtemperature numerical analysis *** turn,the fluid resistance of the fluid network model is corrected according to the results of the 3D fluid-temperature field numerical analysis *** prediction of the temperature distribution of windings is realised by the coupling calculation between the fluid network model and the 3D fluid-temperature field numerical analysis *** on this,the effect of the loading method of the heat source is also investigated using the proposed *** hotspot temperatures of the high-voltage,medium-voltage,and low-voltage windings are 89.43,86.33,and 80.96°C,***,an experimental platform is built to verify the *** maximum relative error between calculated and measured values is 4.42%,which meets the engineering accuracy requirement.

关键词： winding network distribution

来源：评论

学校读者我要写书评

暂无评论

Adaptive Barrier-Lyapunov-Functions Based Control Scheme of Nonlinear Pure-Feedback Systems with Full State Constraints and Asymptotic Tracking Performance

引用

Journal of Systems science & Complexity 2024年第3期37卷 965-984页

作者： NIU Ben WANG Xiaoan WANG Xiaomei WANG Xinjun LI Tao Faculty of Electronic Information and Electrical Engineering Dalian University of TechnologyDalian 116024China School of Information Science and Engineering Shandong Normal UniversityJinan 250014China School of Mathematics Southeast UniversityNanjing 211189China School of Automation Nanjing University of Information Science and TechnologyNanjing 210044China

In this paper,the authors propose an adaptive Barrier-Lyapunov-Functions(BLFs)based control scheme for nonlinear pure-feedback systems with full state *** to the coexist of the non-affine structure and full state constraints,it is very difficult to construct a desired controller for the considered *** to the mean value theorem,the authors transform the pure-feedback system into a system with strict-feedback structure,so that the well-known backstepping method can be ***,in the backstepping design process,the BLFs are employed to avoid the violation of the state constraints,and neural networks(NNs)are directly used to online approximate the unknown packaged nonlinear *** presented controller ensures that all the signals in the closed-loop system are bounded and the tracking error asymptotically converges to ***,it is shown that the constraint requirement on the system will not be violated during the ***,two simulation examples are provided to show the effectiveness of the proposed control scheme.

关键词： Asymptotic tracking control barrier Lyapunov functions full state constraints nonlinear pure-feedback systems

来源：评论

学校读者我要写书评

暂无评论

A consensus time synchronization protocol in wireless sensor network

引用

Journal of Systems engineering and electronics 2023年第6期34卷 1465-1472页

作者： CHEN Kesong ZHANG Yu School of Information and Communication Engineering University of Electronic Science and Technology of ChinaChengdu 611731China

Time synchronization is one of the base techniques in wireless sensor networks(WSNs).This paper proposes a novel time synchronization protocol which is a robust consensusbased algorithm in the existence of transmission delay and packet *** compensates for transmission delay and packet loss firstly,and then,estimates clock skew and clock offset in two *** and experiment results show that the proposed protocol can keep synchronization error below 2μs in the grid network of 10 nodes or the random network of 90 ***,the synchronization accuracy in the proposed protocol can keep constant when the WSN works up to a month.

关键词： time synchronization consensus algorithm transmission delay packet loss wireless sensor network(WSN).

来源：评论

学校读者我要写书评

暂无评论

DRAM:A DRL-based resource allocation scheme for MAR in MEC

引用

Digital Communications and Networks 2023年第3期9卷 723-733页

作者： Tongyu Song Xuebin Tan Jing Ren Wenyu Hu Sheng Wang Shizhong Xu Xiong Wang Gang Sun Hongfang Yu School of Information and Communication Engineering University of Electronic Science and Technology of ChinaChengduPR China

Mobile Edge Computing(MEC)and 5G technology allow clients to access computing resources at the network frontier,which paves the way for applying Mobile Augmented Reality(MAR)*** the MEC paradigm,MAR clients can offload complex tasks to the MEC server and enhance the human perception of the world by merging the received virtual information with the real ***,the resource allocation problem arises as a critical challenge in circumstances where several MAR clients compete for limited resources at the network *** this paper,we aim to design an online resource allocation scheme on the MEC server that takes both high quality of experience and good fairness performance for MAR clients into *** first formulate this problem as a Markov decision process and tackle the challenge of applying the deep reinforcement learning ***,we propose DRAM,a Deep reinforcement learning-based Resource allocation scheme for mobile Augmented reality service in *** also propose a self-adaptive algorithm on the MAR client that is derived based on the analysis of the MAR service to tackle client adaptation *** simulation results demonstrated that DRAM can provide high quality of experience and simultaneously achieve good fairness performance by coordinating with clients’adaptation algorithms.

关键词： Mobile edge computing Mobile augmented reality Deep reinforcement learning Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Faster AMEDA-A Hybrid Mesoscale Eddy Detection Algorithm

引用

Computer Modeling in engineering & sciences 2024年第11期141卷 1827-1846页

作者： Xinchang Zhang Xiaokang Pan Rongjie Zhu Runda Guan Zhongfeng Qiu Biao Song Organization Department Nanjing University of Information Science and TechnologyNanjing210044China School of Computer Science Nanjing University of Information Science and TechnologyNanjing210044China School of Teacher Education Nanjing University of Information Science and TechnologyNanjing210044China School of Electronics&Information Engineering Nanjing University of Information Science and TechnologyNanjing210044China School of Software Nanjing University of Information Science and TechnologyNanjing210044China

Identification of ocean eddies from a large amount of ocean data provided by satellite measurements and numerical simulations is crucial,while the academia has invented many traditional physical methods with accurate detection capability,but their detection computational efficiency is *** recent years,with the increasing application of deep learning in ocean feature detection,many deep learning-based eddy detection models have been developed for more effective eddy detection from ocean *** it is difficult for them to precisely fit some physical features implicit in traditional methods,leading to inaccurate identification of ocean *** this study,to address the low efficiency of traditional physical methods and the low detection accuracy of deep learning models,we propose a solution that combines the target detection model Faster Region with CNN feature(Faster R-CNN)with the traditional dynamic algorithm Angular Momentum Eddy Detection and Tracking Algorithm(AMEDA).We use Faster R-CNN to detect and generate bounding boxes for eddies,allowing AMEDA to detect the eddy center within these bounding boxes,thus reducing the complexity of center *** demonstrate the detection efficiency and accuracy of this model,this paper compares the experimental results with AMEDA and the deep learning-based eddy detection method *** results show that the eddy detection results of this paper are more accurate than eddyNet and have higher execution efficiency than AMEDA.

关键词： Mesoscale eddy detection object detection contour detection remote sensing

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：