检索结果-南通市图书馆

An efficient and balanced feature pyramid network: shared-FPN

Optoelectronics Letters 2025年

作者： Tian Sulong Qin Longfei Pang Wenchao Gao Ying zhao dexin School of Computer Science and Engineering， Tianjin University of Technology Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology， Tianjin University of Technology

In the field of object detection， it is challenging to achieve a balance between neck complexity and accuracy. To address this issue， we propose an efficient and decoupled neck module called Shared-FPN(Shared Feature Pyramid Network). Our design aims to effectively address the trade-off between these two factors. Neck modules are crucial for effectively fusing multi-scale information in neural network models. However， traditional neck designs， with their single structure， have struggled to significantly improve detection performance. In recent years， the incorporation of neural architecture search techniques in neck design has led to an increase in model complexity. This trade-off between model complexity and detection performance highlights the need for more efficient neck design approaches. Not only does Shared-FPN not increase the number of model parameters and FLOPs， but it also can be easily ported to any of the detection models. It improves on PAFPN by using transposed convolution with a large convolution kernel as the upsampling module， designing spatial pyramidal pooling-fast downsampling(SPPFD) based on shared pooling， and designing shared convolution as the right part module. To evaluate the performance of Shared-FPN in object detection tasks， we conducted experiments on object detection datasets. The results show that Shared-FPN achieved excellent performance across all sizes. In particular， on the VOC 2012 dataset， which contains mostly large objects， Shared-FPN's mAP was improved by 11.2% compared to FPN with YOLOX-s as the detector. On the COCO dataset， which contains mostly small objects， Shared-FPN's mAP was improved by 7.8% compared to FPN and 7.1% compared to PAFPN with Faster RCNN as the detector. Shared-FPN can be easily inserted into any of the detectors for better performance in various scenarios， such as small， medium or large objects.

关键词：

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Shallow water source localization using a mobile short horizontal array

引用

Journal of Systems Engineering and Electronics 2013年第5期24卷 749-760页

作者： dexin zhao Woojae Seong Keunhwa Lee Zhiping Huang Department of Instrument Science and Technology College of Mechatronics and AutomationNational University of Defense Technology Department of Naval Architecture and Ocean Engineering College of EngineeringSeoul National University Research Institute and Marine System Engineering Seoul National University

This paper presents an approach to the challenging is- sue of passive source localization in shallow water using a mobile short horizontal linear array with length less than ten meters. The short array can be conveniently placed on autonomous underwa- ter vehicles and deployed for adaptive spatial sampling. However, the use of such small aperture passive sonar systems makes it difficult to acquire sufficient spatial gain for localizing long-range sources. To meet the requirement, a localization approach that employs matched-field based techniques that enable the short ho- rizontal linear array is used to passively localize acoustic sources in shallow water. Furthermore, the broadband processing and inter-position processing provide robustness against ocean en- vironmental mismatch and enhance the stability of the estimation process. The proposed approach＇s ability to localize acoustic sources in shallow water at different signal-to-noise ratios is examined through the synthetic test cases where the sources are located at the endfire and some other bearing of the mobile short horizontal linear array. The presented results demonstrate that the positional parameters of the estimated source build up over time as the array moves at a low speed along a straight line at a constant depth.

关键词： passive source localization shallow water short horizontal array matched-field based technique broadband process- ing inter-position processing.

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Effects of dynamic recrystallization and strain-induced dynamic precipitation on the corrosion behavior of partially recrystallized Mg-9Al-lZn alloys

引用

Journal of Magnesium and Alloys 2020年第4期8卷 1016-1037页

作者： Yenny Cubides dexin zhao Lucas Nash Digvijay Yadav Kelvin Xie Ibrahim Karaman Homero Castaneda Department of Materials Science and Engineering Texas A&M University575 Ross StreetCollege StationTX 77840USA Department of Engineering Technology and Industrial Distribution 106 Ross StreetCollege StationTX 77843USA National Corrosion and Materials Reliability Laboratory 1041 RELLIS PkwyBryanTX 77807USA

The corrosion susceptibility of recrystallized and un-recrystallized grains in equal channel angular pressed(ECAPed)Mg-9Al-lZn(AZ91)alloys immersed in chloride containing media was investigated through immersion testing and an electrochemical microcell technique coupledwith high resolution techniques such as scanning Kelvin probe force microscopy(SKPFM),transmission electron microscopy(TEM),andelectron backscatter diffraction(EBSD).During ECAP,dynamic recrystallization(DRX)and strain-induced dynamic precipitation(SIDP)simultaneously occurred,resulting in a bimodal grain structure of original elongated coarse grains and newly formed equiaxed fine grainswith a large volume fraction ofβ-Mg17Al12 *** preferentially initiates and propagates in the DRXed grains,owing tothe greater microchemistry difference between theβ-Mg17Al12 precipitates formed at the DRXed grain boundaries and the adjacentα-Mgmatrix,which induces a strong microgalvanic coupling between these ***,the weaker basal texture of the DRXed grainsalso makes these grains more susceptible to electrochemical reactions than the highly textured un-DRXed *** influence of dynamicrecrystallization and dynamic precipitation was also studied in ECAPed alloys with differenl levels of deformation strain through corrosion andelectrochemical *** the strain level led to a more uniform corrosion with a shallow penetration depth,lower corrosion ratevalues,and higher protective ability of the oxide ***,higher levels of strain resulted in greater hardness values of the *** superior corrosion resistance and strength of the ECAPed alloys with increasing strain level was attributed to the combination ofsmaller DRXed grain size,higher DRX ratio,and higher volume fraction of uniformly distributed fineβ-Mg17Al12 precipitates.c 2020 Published by Elsevier *** behalf of Chongqing University.

关键词： Magnesium alloy Bimodal grain structure Dynamic recrystallization Dynamic precipitation Severe plastic deformation Microgalvanic coupling

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

A lightweight convolutional neural network for large-scale Chinese image caption

引用

Optoelectronics Letters 2021年第6期17卷 361-366页

作者： zhao dexin YANG Ruixue GUO Shutao Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology Tianjin University of TechnologyTianjin 300384China

Image caption is a high-level task in the area of image understanding,in which most of the models adopt a convolutional neural network(CNN)to extract image features assigning a recurrent neural network(RNN)to generate *** tend to design complex networks with deeper layers to improve the performance of feature extraction in recent *** the size of the network could obtain features of high quality,but it is not an efficient way in terms of computational cost.A large number of parameters brought by CNN makes the research difficult to apply in human daily *** order to reduce the information loss of the convolutional process with less cost,we propose a lightweight convolutional neural network,named as Bifurcate-CNN(B-CNN).Furthermore,recent works are devoted to generating captions in English,in this paper,we develop an image caption model that generates descriptions in *** with Inception-v3,the depth of our model is shallower with fewer parameters,and the computational cost is *** on the AI CHALLENGER dataset,we prove that our model can enhance the performance,improving BLEU-4 from 46.1 to 49.9 and CIDEr from 142.5 to 156.6 respectively.

关键词： neural network image

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

An adaptive particle filter （APF） for estimating states of nonlinear system with unknown parameters

An adaptive particle filter （APF） for estimating states of...

引用

第25届中国控制与决策会议

作者： dexin zhao Ting Li Zhiping Huang Shaojing Su Department of Instrument Science and Technology College of Mechatronics and Automation National University of Defense Technology

ISBN: (纸本)9781467355339

This paper presents an adaptive particle filter （APF） for estimating the states of the nonlinear system with unknown parameters. To implement this algorithm, the process noise covariance is adjusted according to the innovation-based adaptive estimation （IAE） approach, which only takes advantage of one state model without the requirement of knowing the parameters. In addition, the Maximum-A-Posterior （MAP） method is used to estimate the parameters to improving the performance of APF. Therefore, the APF can modify the prior distribution of the particles, and then alleviate the sample degeneracy problem which is common in particle filter （PF）. Numerical simulations of Logistic map are conducted to demonstrate the effectiveness of our proposed method.

关键词： Particle filter IAE Nonlinear system State estimation Unknown parameters

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

A Secure Device Management Scheme with Audio-Based Location Distinction in IoT

引用

Computer Modeling in Engineering & Sciences 2024年第1期138卷 939-956页

作者： Haifeng Lin Xiangfeng Liu Chen Chen Zhibo Liu dexin zhao Yiwen Zhang Weizhuang Li Mingsheng Cao College of Economics and Management Nanjing University of Aeronautics and AstronauticsNanjing211106China Chengdu Aircraft Industrial(Group)Co. Ltd.Chengdu610073China Academy of Military Sciences of PLA Beijing100091China School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu610054China Ningbo Web King Technology Joint Stock Co. Ltd.Ningbo315000China The Network and Data Security Key Laboratory of Sichuan Province University of Electronic Science and Technology of ChinaChengdu610054China

Identifying a device and detecting a change in its position is critical for secure devices management in the Internet of Things(IoT).In this paper,a device management system is proposed to track the devices by using audio-based location distinction *** the proposed scheme,traditional cryptographic techniques,such as symmetric encryption algorithm,RSA-based signcryption scheme,and audio-based secure transmission,are utilized to provide authentication,non-repudiation,and confidentiality in the information interaction of the management ***,an audio-based location distinction method is designed to detect the position change of the ***,the audio frequency response(AFR)of several frequency points is utilized as a device *** device signature has the features as follows.(1)Hardware Signature:different pairs of speaker and microphone have different signatures;(2)Distance Signature:in the same direction,the signatures are different at different distances;and(3)Direction Signature:at the same distance,the signatures are different in different *** on the features above,amovement detection algorithmfor device identification and location distinction is ***,a secure communication protocol is also proposed by using traditional cryptographic techniques to provide integrity,authentication,and non-repudiation in the process of information interaction between devices,Access Points(APs),and *** experiments are conducted to evaluate the performance of the proposed *** experimental results show that the proposedmethod has a good performance in accuracy and energy consumption.

关键词： Acoustic hardware fingerprinting device management IoT location distinction

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

中国市场经济的由来——市场关系发展的三个阶段

引用

中南财经政法大学学报 2010年第2期 77-82,116页

作者：赵德馨中南财经政法大学经济学院湖北武汉430073

本文依次考察了中国氏族社会后期以来4000多年市场关系的发展所经历的商品货币关系、商品经济与市场经济三个阶段,以及各个阶段中市场关系诸要素的演变过程与特征,证明它们在不断地向前发展;市场具有为自己扫除前进障碍的内在力量;市场... 详细信息

本文依次考察了中国氏族社会后期以来4000多年市场关系的发展所经历的商品货币关系、商品经济与市场经济三个阶段,以及各个阶段中市场关系诸要素的演变过程与特征,证明它们在不断地向前发展;市场具有为自己扫除前进障碍的内在力量;市场经济向前发展是不以人的意志为转移的历史规律;中国今天的市场经济是经济运行4000多年的必然结局。

关键词：商品经济市场关系市场经济

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

傅筑夫对经济史学科的探索及其独特贡献

引用

中国经济史研究 2023年第2期 181-188页

作者：赵德馨中南财经政法大学经济学院 430073

傅筑夫从读大学预科时起,学习兴趣从理科转到文科中的国学、文学、哲学,再转向社会科学,在而立之年确定以中国经济史学为志业,先博而后专。他是中国最早在大学讲授经济史学课程的教授之一,也是现代型中国经济史学科的奠基者之一。他治... 详细信息

傅筑夫从读大学预科时起,学习兴趣从理科转到文科中的国学、文学、哲学,再转向社会科学,在而立之年确定以中国经济史学为志业,先博而后专。他是中国最早在大学讲授经济史学课程的教授之一,也是现代型中国经济史学科的奠基者之一。他治中国经济史,从殷商到近代;治欧洲经济史,从罗马帝国的经济到工业革命的完成,可谓博通古今,学贯中西。他在中西比较中发现,早在2000多年前,中国经济就走上了一条与欧洲不同的发展道路。他的论著覆盖经济史学科体系中的几个主要分支学科。他的观点自成一家。这种情况,在他们那一代经济史学人中是少有的。

关键词：傅筑夫经济史学科中国经济史欧洲经济史

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

1949～2002年:走向共同富裕的两条思路及其实践经验

引用

当代中国史研究 2007年第2期14卷 29-38页

作者：赵德馨中南财经政法大学经济学院

1949年中国共产党执政以后,在领导中国人民奔向共同富裕目标的进程中,经历了1978年之前和之后的两个阶段。两个阶段都成功地解决了特定历史阶段面临的主要任务,也都留下了新问题。实践表明:在解决共同富裕的模式问题上,指导思想起决定作... 详细信息

1949年中国共产党执政以后,在领导中国人民奔向共同富裕目标的进程中,经历了1978年之前和之后的两个阶段。两个阶段都成功地解决了特定历史阶段面临的主要任务,也都留下了新问题。实践表明:在解决共同富裕的模式问题上,指导思想起决定作用;平均主义是共同富裕的对立物,两极分化也是共同富裕的对立物;找到既有效率又富公平的分配方式,是一个社会主义难题。正确地总结正面与反面的经验,有助于解决这个难题,并将实践推向新阶段。

关键词： 1949-2002年共同富裕思路经验

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Enhancement Channel Estimation Using Outer-Product Decomposition Algorithm Based on Frequency Transformation

引用

Journal of Marine Science and Application 2020年第2期19卷 283-292页

作者： Xiukun Li Ji Wang dexin zhao Acoustic Science and Technology Laboratory Harbin Engineering UniversityHarbin 150001China Key Laboratory of Marine Information Acquisition and Security(Harbin Engineering University) Ministry of Industry and Information TechnologyHarbin 150001China College of Underwater Acoustic Engineering Harbin Engineering UniversityHarbin 150001China Advanced Interdisciplinary Technology Research Center National Innovation Institute of Defense TechnologyBeijing 100010China

The outer-product decomposition algorithm(OPDA)performs well at blindly identifying system ***,the direct use of the OPDA in systems using bandpass source will lead to *** study proposes an approach to enhance the channel estimation quality of a bandpass source that uses *** approach performs frequency domain transformation on the received signal and obtains the optimal transformation parameter by minimizing the p-norm of an error ***,the proposed approach extends the application of OPDA from a white source to a bandpass white source or chirp *** formulas and simulation results show that the proposed approach not only reduces the estimation error but also accelerates the algorithm in a bandpass system,thus being highly feasible in practical blind system identification applications.

关键词： Blind identification Outer-product decomposition algorithm Bandpass white signal Chirp signal Second-order statistics

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：