检索结果-内蒙古大学图书馆

A data augmentation method for war trauma based on the regression model and deep neural networks

Soft Computing 2024年第23期28卷 13527-13540页

作者： Yin, Jibin Liao, Gongshan Zhang, Yi Sun, Peiyu Guo, Sisi Key Laboratory of Computer Science and Technology Kunming University of Science and Technology Kunming650500 China Naval Medical Service Training Base Naval Medical University Shanghai200433 China School of Health Science and Engineering University of Shanghai for Science and Technology Shanghai200093 China

War trauma data, a core element of military research, have been attracting significant attention from several countries, despite limited availability. In this study, we propose a data augmentation method for war trauma that combines a trauma scoring system and deep learning to generate large-scale and effective synthetic data. First, the Abbreviated War Trauma Scale (AWTS), including the trauma scoring table and the vital signs scoring table, was developed by collecting information on battlefield environment injuries. Second, a new war trauma severity scoring (NWTSS) equation evaluating injury severity was proposed using multiple linear regression analysis on manually labelled synthetic samples. Then, based on the AWTS and NWTSS, we formulated a synthetic algorithm for generating large-scale automatically labelled cases. Finally, to verify the effectiveness of the synthetic data, we trained deep-learning classifiers with these data and tested them using real data. The experimental results show that the classifiers performed best when the size of the synthetic data scale was around 50,000. Furthermore, the text classification model designed in this study for war trauma data achieved 85.59% accuracy, outperforming the current popular text classifiers, indicating that the synthetic data effectively reflected the real situation. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词： Multiple linear regression

来源：评论

学校读者我要写书评

暂无评论

FDCPNet:feature discrimination and context propagation network for 3D shape representation

引用

虚拟现实与智能硬件(中英文) 2025年第1期7卷 83-94页

作者： Weimin SHI Yuan XIONG Qianwen WANG Han JIANG Zhong ZHOU State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and EngineeringBeihang UniversityBeijing 100191China

Background Three-dimensional(3D)shape representation using mesh data is essential in various applications,such as virtual reality and simulation *** methods for extracting features from mesh edges or faces struggle with complex 3D models because edge-based approaches miss global contexts and face-based methods overlook variations in adjacent areas,which affects the overall *** address these issues,we propose the Feature Discrimination and Context Propagation Network(FDCPNet),which is a novel approach that synergistically integrates local and global features in mesh *** FDCPNet is composed of two modules:(1)the Feature Discrimination Module,which employs an attention mechanism to enhance the identification of key local features,and(2)the Context Propagation Module,which enriches key local features by integrating global contextual information,thereby facilitating a more detailed and comprehensive representation of crucial areas within the mesh *** Experiments on popular datasets validated the effectiveness of FDCPNet,showing an improvement in the classification accuracy over the baseline ***,even with reduced mesh face numbers and limited training data,FDCPNet achieved promising results,demonstrating its robustness in scenarios of variable complexity.

关键词： 3D shape representation Mesh model MeshNet Feature discrimination Context propagation

来源：评论

学校读者我要写书评

暂无评论

Loop Subgraph-Level Greedy Mapping Algorithm for Grid Coarse-Grained Reconfigurable Array

引用

Tsinghua science and technology 2023年第2期28卷 330-343页

作者： Naijin Chen Fei Cheng Chenghao Han Jianhui Jiang Xiaoqing Wen School of Computer and Information Science Anhui Polytechnic UniversityWuhu 241000China School of Software Engineering Tongji UniversityShanghai 201804China Department of Computer Science and Networks Kyushu Institute of TechnologyFukuoka 820-8502Japan

To solve the problem of grid coarse-grained reconfigurable array task mapping under multiple constraints,we propose a Loop Subgraph-Level Greedy Mapping(LSLGM)algorithm using parallelism and processing element *** the constraint of a reconfigurable array,the LSLGM algorithm schedules node from a ready queue to the current reconfigurable cell array *** mapping a node,its successor’s indegree value will be dynamically *** its successor’s indegree is zero,it will be directly scheduled to the ready queue;otherwise,the predecessor must be dynamically *** the predecessor cannot be mapped,it will be scheduled to a blocking *** dynamically adjust the ready node scheduling order,the scheduling function is constructed by exploiting factors,such as node number,node level,and node *** with the loop subgraph-level mapping algorithm,experimental results show that the total cycles of the LSLGM algorithm decreases by an average of 33.0%(PEA44)and 33.9%(PEA_(7×7)).Compared with the epimorphism map algorithm,the total cycles of the LSLGM algorithm decrease by an average of 38.1%(PEA_(4×4))and 39.0%(PEA_(7×7)).The feasibility of LSLGM is verified.

关键词： Grid Coarse-Grained Reconfigurable Array(GCGRA) mapping loop subgraph scheduling

来源：评论

学校读者我要写书评

暂无评论

Asynchronous Energy-to-peak Control for 2D Roesser-type Markov Jump Systems

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2024年第11期51卷 1761-1768页

作者： Chen, Qing Zhang, Yu Han, Chengyi Chen, Ling Zhou, Jianping The School of Computer Science and Technology Anhui University of Technology Ma’anshan243032 China The School of Electrical and Information Engineering Wanjiang University of Technology Ma’anshan243032 China

This paper studies asynchronous energy-to-peak control for 2D Roesser-type Markov jump systems (RTMJSs). Given the practical challenge of obtaining the system state, output-feedback is utilized for closing the control loop instead of state feedback. The asynchronous behavior between the controlled plant and the controller is represented by a hidden Markov model. A sufficient condition for the stochastic stability and energy-to-peak disturbance attenuation (EPDA) of the closed-loop RTMJS is established using a Lyapunov function and Schur’s complement. Under fixed conditional probability, a scheme for asynchronous output-feedback controller design is proposed, employing variable replacement and matrix transformation to cope with the nonlinear terms. The present work is then expanded to the uncertain conditional probability scenario and a related controller design method is developed. Finally, the effectiveness of the proposed design methods is demonstrated through a thermal process. © (2024), (International Association of Engineers). All rights reserved.

关键词： Lyapunov functions

来源：评论

学校读者我要写书评

暂无评论

P3DC:Reducing DRAM Cache Hit Latency by Hybrid Mappings

引用

Journal of computer science & technology 2024年第6期39卷 1341-1360页

作者： Ye Chi Ren-Tong Guo Xiao-Fei Liao Hai-Kun Liu Jianhui Yue National Engineering Research Center for Big Data Technology and System Wuhan 430074China Services Computing Technology and System Laboratory Wuhan 430074China Cluster and Grid Computing Laboratory Wuhan 430074China School of Computer Science and Technology Huazhong University of Science and TechnologyWuhan 430074China School of Big Data and Internet Shenzhen Technology UniversityShenzhen 518118China Department of Computer Science Michigan Technological UniversityHoughton 49931-1295U.S.A.

Die-stacked dynamic random access memory(DRAM)caches are increasingly advocated to bridge the performance gap between the on-chip cache and the main *** fully realize their potential,it is essential to improve DRAM cache hit rate and lower its cache hit *** order to take advantage of the high hit-rate of set-association and the low hit latency of direct-mapping at the same time,we propose a partial direct-mapped die-stacked DRAM cache called *** design is motivated by a key observation,i.e.,applying a unified mapping policy to different types of blocks cannot achieve a high cache hit rate and low hit latency *** address this problem,P3DC classifies data blocks into leading blocks and following blocks,and places them at static positions and dynamic positions,respectively,in a unified set-associative *** also propose a replacement policy to balance the miss penalty and the temporal locality of different *** addition,P3DC provides a policy to mitigate cache thrashing due to block type *** results demonstrate that P3DC can reduce the cache hit latency by 20.5%while achieving a similar cache hit rate compared with typical set-associative caches.P3DC improves the instructions per cycle(IPC)by up to 66%(12%on average)compared with the state-of-the-art direct-mapped cache—BEAR,and by up to 19%(6%on average)compared with the tag-data decoupled set-associative cache—DEC-A8.

关键词： die-stacked dynamic random access memory(DRAM) cache set-associative direct-mapped hit latency

来源：评论

学校读者我要写书评

暂无评论

Intergranular deformation mechanism stimulated by{10-12}extension twins in AZ81-La alloy

引用

Journal of Rare Earths 2024年第12期42卷 2325-2331,I0005页

作者： Biwu Zhu Jing Su Xiao Liu Changping Tang Wenhui Liu Luoxing Li Congchang Xu Pengcheng Guo College of Marine Equipment and Mechanical Engineering Jimei UniversityXiamen 361021China Research Institute of HNU in Chongqing Hunan UniversityChongqing401135China Department of Mining and Materials Engineering McGill University3610 Rue UniversityMontrealQuebec H3A 0C5Canada School of Materials Science and Engineering Hunan University of Science and TechnologyXiangtan411201China

In polycrystalline Mg-RE magnesium alloys,the lattice parameter,dislocation dynamics and critical resolved shear stress(CRSS)for various deformation modes are altered with the addition of RE atom,finally affecting deformation mode and strain accommodation mechanism among neighboring *** compression tests were performed on as-extruded AZ81-La Mg alloy samples with the c-axis of the majority of crystals vertical to the compression *** variants and dominant slip systems were examined by electron backscattered diffraction(EBSD)*** is found by in-grain misoriention axis(IGMA)analysis that the plastic deformation is mainly accommodated by a combination of the pyramidal slip and the basal *** dominant twin variant is{0112}<0111>and{1102}<1101>.Here,we applied a modified displacement gradient accommodation(m-DGA)method to evaluate the selection mechanism of twin/slip in neighbour grain stimulated by{1012}extension *** is found that the activated slip system and/or twin variant in the neighbouring grain is determined by the accommodation of the major shear strain induced by{1012}extension twin in such a way that strain concentration along grain boundaries can be relaxed.

关键词： AZ81-La alloy Twinning Slip Accommodation Rotation tensor Rare earths

来源：评论

学校读者我要写书评

暂无评论

LMSA:Lightweight Multi-Scale Attention Model for Garbage Detection Based on YOLOV5 24

LMSA:Lightweight Multi-Scale Attention Model for Garbage Det...

引用

24th IEEE International Conference on Software Quality, Reliability and Security Companion, QRS-C 2024

作者： Liu, Fumin Wu, Yuezhong Huang, Haiyang Shen, Xuehao Deng, Jiefu College of Computer Science Hunan University of Technology Hunan Zhuzhou China College of Railway Transportation Hunan University of Technology Hunan Zhuzhou China School of Computer Science and Technology Hainan University Hainan Haikou China School of Foreign Languages and Cultures Xihua University Sichuan Chengdu China

ISBN: (纸本)9798350365658

The challenges of small target detection, inference speed, and detection accuracy are key research areas in intelligent waste classification recognition, particularly the incorporation of attention mechanisms into algorithms. Existing self-attention mechanisms often face speed limitations due to low efficiency, particularly in MHSA's tensor reshaping and element-by-element computation, which are computationally intensive and challenging for embedded device deployment. To address these issues, this paper introduces a Lightweight Multi-Scale Attention Model (LMSA) based on YOLOV5 and the self-attention mechanism. The model constructs a new lightweight multi-scale attention module, utilizing a single memory-constrained MHSA between effective FFN layers to enhance channel communication while improving access efficiency. Additionally, a cascaded grouped attention module is proposed to provide different divisions of full features to the attention head, saving computational overhead and enhancing attention diversity. Experimental results demonstrate that on the VisDrone2021 dataset, the AP75 of LMSA is 41.43%, achieving the state-of-the-art (SOTA) among mainstream models. On a self-constructed dataset, the AP50 of LMSA reaches 74.55%. With equivalent accuracy, LMSA has lower computational power consumption, and the inference speed reaches 140fps on an RTX3090 graphics card. The model's advantages include fewer parameters, high accuracy, and fast inference speed, making it more suitable for embedded detection equipment and contributing to the intelligent development of waste detection. © 2024 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Ambient-condition strategy for rapid mass production of crystalline gallium oxide nanoarchitectures toward device application

引用

Journal of Materials science & technology 2023年第32期163卷 150-157页

作者： Dongdong Zhang Hao Yu Guiqin You Gang Shao Zhi Fang Zhao Liang Tian Zhang Huilin Hou Lin Wang Qiaochu Chen Jie Teng Weiyou Yang College of Materials Science and Engineering Hunan UniversityChangsha 410082China Institute of Micro/Nano Materials and Devices Ningbo University of TechnologyNingbo 315211China School of Materials Science and Engineering Zhengzhou UniversityZhengzhou 450001China

Currently,the synthesis of nanostructures still encounters two grand challenges:one is the often-required high temperatures and/or high pressures,and the other is the scalable ***,to break through such bottlenecks,we demonstrate an ambient-condition strategy for rapid mass production of fourthgeneration semiconductor Ga_(2)O_(3)nanoarchitectures assembled by single-crystalline nanosheets in a controlled manner based on *** growth is fundamentally determined by the introduced ethanediamine in rationally designed source materials,which could not only protect the metal Ga against reacting with H_(2)O into GaOOH,but facilitate the reaction of Ga with OH·radicals for target crystalline Ga_(2)O_(3)*** a proof of concept for applications,the as-fabricated Ga_(2)O_(3)nanoarchitectures exhibit superb performances for electromagnetic wave absorption with a reflection loss value of 52.2 dB at 8.1 GHz,and ammonia sensing with high sensitivity and selectivity at room temperature,representing their bright future to be commercially applied in modern devices.

关键词： Semiconductor Crystalline nanostructures Crystal growth Ambient condition Device application

来源：评论

学校读者我要写书评

暂无评论

Extending Scratch Framework to Improve the Analytical Skills of Undergraduate Students 24

Extending Scratch Framework to Improve the Analytical Skills...

引用

10th International Conference on computer technology Applications, ICCTA 2024

作者： Akbar, Ubaid Ul Akbar, Saeed Ali, Mumtaz Ullah, Rahmat Khan, Rizwan Lopes, Ivandro Ortet Department of Computer Science City University of Science and Information Technology Peshawar Pakistan School of Computer Science and Technology Zhejiang Normal University Jinhua321004 China School of Computer Science and Electronic Engineering University of Essex United Kingdom School of Cyber Science & Engineering Huazhong University of Science & Technology Wuhan China

ISBN: (纸本)9798400716386

Scratch is an innovative and the most popular block-based Visual Programming Language designed for beginners to learn programming effectively. However, it lacks the capacity to allow users to learn and solve larger real-world problems such as the Travelling Salesman Problem (TSP). Hence, there is a need for an accessible and effective tool to assist beginners and novice programmers in learning and tackling the TSP. This paper introduces ScratchTSP, an extension of the Scratch programming Framework, specifically designed for the TSP. It offers an opportunity for beginners and novice programmers to grasp the TSP and the various algorithms used to solve it. For evaluation, we conduct a survey among undergraduate students to assess the usability and usefulness of the proposed ScratchTSP extension. Survey findings indicate that ScratchTSP offers a user-friendly and effective tool for beginners or novice programmers, aiding them in learning the TSP while developing their critical thinking and problem-solving skills. © 2024 ACM.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Text region detection algorithm for medical instruments screen based on improved DBNet 2

Text region detection algorithm for medical instruments scre...

引用

2nd International Conference on Pattern Recognition, Machine Vision and Intelligent Algorithms, PRMVIA 2024

作者： Liu, Jiaming Chen, Wei Heilongjiang University of Science and Technology School of Computer and Information Engineering Nantong China Heilongjiang University of Science and Technology School of Computer and Information Engineering Harbin China

ISBN: (纸本)9798350389746

Recognition of text in images has become increasingly important with the rapid development of cameras, and the main technology involved is Optical Character Recognition (OCR). In order to enhance the recognition accuracy of traditional text region detection algorithms for medical instrument screen data, which contains irregular and small objects, this paper introduces an improved scheme incorporating Adaptive Spatial Feature Fusion (ASFF) and Gaussian Error Linear Unit (GELU) activation function, and proposes the DBNeti model. The introduction of ASFF aims to address the limitations of fixed feature fusion in traditional text recognition algorithms and to enhance the performance of target detectors at different scales. The GELU is introduced as the activation function of the DBNet algorithm, utilizing its nonlinear characteristics, gradient preservation, and computational efficiency to enhance the expressiveness and generalization ability of the model. Experiments demonstrate that the improved model, DBNeti, has achieved optimal performance across all evaluation indices, including an Accuracy of 0.8189, a Recall of 0.3633, a Precision of 0.6858, and an F-measure of 0.4750. Compared to DBNet, DBNeti improves the Accuracy, Recall, Precision, and F-measure indices by 3.3%, 6.1%, 3.1%, and 4.2%, respectively. This study demonstrates the superiority of the DBNeti algorithm in the text recognition task. © 2024 IEEE.

关键词： Image segmentation

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：