检索结果-南通市图书馆

UAV image target localization method based on outlier filter and frame buffer

Chinese Journal of Aeronautics 2024年第7期37卷 375-390页

作者： Yang WANG Hongguang LI Xinjun LI Zhipeng WANG baochang zhang Institute of Unmanned System Beihang UniversityBeijing 100191China School of Electronics and Information Engineering Beihang UniversityBeijing 100191China Institute of Artificial Intelligence Beihang UniversityBeijing 100191China

With rapid development of UAV technology,research on UAV image analysis has gained *** the existing techniques of UAV target localization often rely on additional equipment,a method of UAV target localization based on depth estimation has been ***,the unique perspective of UAVs poses challenges such as the significant field of view variations and the presence of dynamic objects in the *** a result,the existing methods of depth estimation and scale recovery cannot be directly applied to UAV ***,there is a scarcity of depth estimation datasets tailored for UAV perspectives,which makes supervised algorithms *** address these issues,an outlier filter is introduced to enhance the applicability of depth estimation networks to target localization.A frame buffer method is proposed to achieve more accurate scale recovery,so as to handle complex scene textures in UAV *** proposed method demonstrates a 14.29%improvement over the *** with the average recovery results from UAV perspectives,the difference is only 0.88%,approaching the performance of scale recovery using ground truth ***,to overcome the limited availability of traditional UAV depth datasets,a method for generating depth labels from video sequences is *** to state-of-the-art methods,the proposed approach achieves higher accuracy in depth estimation and stands for the first attempt at target localization using image *** algorithm and dataset are available at https://***/uav-tan/uav-object-localization.

关键词： Object localization Deep learning Depth estimate Scale recovery Unmanned Aerial Vehicle(UAV)

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

Multi-block SSD based on small object detection for UAV railway scene surveillance

引用

Chinese Journal of Aeronautics 2020年第6期33卷 1747-1755页

作者： Yundong LI Han DONG Hongguang LI Xueyan zhang baochang zhang Zhifeng XIAO North China University of Technology Beijing 100144China Key Laboratory of Large Structure Health Monitoring and Control Shijiazhuang 050043China Unmanned System Research Institute Beihang UniversityBeijing 100083China Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institutes of Advanced TechnologyChinese Academy of SciencesShenzhen 518055China School of Automation Science and Electrical Engineering Beihang UniversityBeijing 100083China State Key Laboratory of Information Engineering in Surveying Wuhan UniversityWuhan 430079China

A method of multi-block Single Shot Multi Box Detector(SSD)based on small object detection is proposed to the railway scene of unmanned aerial vehicle *** address the limitation of small object detection,a multi-block SSD mechanism,which consists of three steps,is ***,the original input images are segmented into several overlapped ***,each patch is separately fed into an SSD to detect the ***,the patches are merged together through two *** the first stage,the truncated object of the sub-layer detection result is *** the second stage,a sub-layer suppression and filtering algorithm applying the concept of non-maximum suppression is utilized to remove the overlapped boxes of *** boxes that are not detected in the main-layer are *** addition,no sufficient labeled training samples of railway circumstance are available,thereby hindering the deployment of SSD.A two-stage training strategy leveraging to transfer learning is adopted to solve this *** deep learning model is preliminarily trained using labeled data of numerous auxiliaries,and then it is refined using only a few samples of railway scene.A railway spot in China,which is easily damaged by landslides,is investigated as a case *** results show that the proposed multi-block SSD method produces an overall accuracy of 96.6%and obtains an improvement of up to 9.2%compared with the traditional SSD.

关键词： Deep learning Multi-block Single Shot MultiBox Detector(SSD) Objection detection Railway scene Unmanned aerial vehicle remote sensing

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

Optimization of bits allocation and path planning with trajectory constraint in UAV-enabled mobile edge computing system

引用

Chinese Journal of Aeronautics 2020年第10期33卷 2716-2727页

作者： Yizhe LUO Wenrui DING baochang zhang Wenqian HUANG Chunhui LIU School of Electronics and Information Engineering Beihang UniversityBeijing 100083China Institute of Unmanned System Beihang UniversityBeijing 100083China School of Automation Science and Electrical Engineering Beihang UniversityBeijing 100083China

In this paper,an Unmanned Aerial Vehicle(UAV)enabled Mobile Edge Computing(MEC)system is studied,in which UAV acts as server to offer computing offloading service to the Mobile Users(MUs)with limited computing capability and energy *** aim to minimize the total energy consumption of MUs by jointly optimizing the bit allocation for uplink,computing at the UAV and downlink,along with the UAV trajectory in a unified *** this end,a trajectory constraint model is employed to avoid sudden changes of velocity and acceleration during *** to high-order information in use,we lead to a more reasonable nonconvex optimization problem than prior *** Alternating Direction Method of Multipliers(ADMM)method is introduced to solve the optimization problem,which is decomposed into a set of easy subproblems,to meet the requirement on the efficiency in edge *** results demonstrate that our approach leads a smoother UAV trajectory,significantly save the energy consumption for UAV during flying.

关键词： Constraint implementation Edge computing Energy consumption Optimization methodology Unmanned Aerial Vehicles(UAV)

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

Bandit neural architecture search based on performance evaluation for operation selection

引用

Science China(Technological Sciences) 2023年第2期66卷 481-488页

作者： zhang Jian GONG Xuan LIU YuXiao WANG Wei WANG Lei zhang baochang School of Automation Science and Electrical Engineering Beihang UniversityBeijing 100191China School of Computing University at BuffaloNew York 14260USA

Neural architecture search(NAS)plays an important role in many computer vision ***,the high computational cost of forward and backward propagations during the search,process restricts its practical *** this paper,we present the search process as a multi-armed bandit problem,where we take into account the uncertainty caused by the contradiction between the huge search space and limited number of *** NAS optimizes the trade-off between exploitation and exploration for a highly efficient ***,we sampled from a set of operations in one trial,where each operation was weighted by its trial performance and a bias to allow operations with less training to be *** further reduced the search space by abandoning the operation with the lowest potential,significantly reducing the search *** results on the CIFAR-10 dataset show that the resulting architecture achieves the most advanced precision with a search speed approximately two times faster than that of partially connected differentialble architecture *** ImageNet,we attained the most advanced top-1 accuracy of 75.3%with a search time of 1.8 GPU days.

关键词： bandit NAS DARTS upper confidence bounds

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

Chemical Synthesis of Proteins Containing 300 Amino Acids

引用

Chemical Research in Chinese Universities 2020年第5期36卷 733-747页

作者： zhang baochang LI Yulei SHI Weiwei WANG Tongyue zhang Feng LIU Lei Tsinghua-Peking Center for Life Sciences Key Laboratory of Bioorganic Phosphorus Chemistry and Chemical Biology(Ministry of Education)Center for Synthetic and Systems BiologyState Key Laboratory of Chemical Oncogenomics(Shenzhen)Department of ChemistryTsinghua UniversityBeijing 100084P.R.China

Chemical synthesis of proteins containing up to 300 amino acids may cover 30%—50%of all the proteins encountered in biomedical studies and may provide an alternate approach to the usually used recombinant expression teclmology,vastly expanding the chemical space of the *** the present review article,we tried to survey the recent progresses made for more rapid synthesis of increasingly long peptides and more efficient ligation of multiple peptide *** developments of seminal methods by many research groups have greatly contributed to the recent breakthroughs in the successful total synthesis of a number of functionally important proteins,such as oligoubiquitins,bacterial GroEL/ES chaperones,and mirror-image DNA *** these studies,a potential bottleneck has also been recognized for the chemical synthesis of large proteins,namely,how to ensure that each peptide segment from a large protein avoids unfavorable aggregation when dissolved in aqueous *** new methods,such as removable backbone modification(RBM)strategy have been developed to overcome this bottleneck,while more studies need to be carried out to develop more effective and less costly methods that ultimately,may lead to fully automatable chemical synthesis of customized proteins of 300 amino acids bearing any artificial designs.

关键词： Chemical protein synthesis Solid-phase synthesis Peptide Ligation 300 amino acid

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

Chemical synthesis and racemic crystallization of rat C5a-desArg

引用

Chinese Chemical Letters 2020年第3期31卷 693-696页

作者： Chao Zuo baochang zhang Meng Wu Donald Bierer Jing Shi Ge-Min Fang School of Life Science Institutes of Physical Science and InformationTechnologyAnhui UniversityHefei 230601China Department of Chemistry Tsinghua UniversityBeijing 100084China Department of Chemistry University of Science and Technology of ChinaHefei 230601China Bayer AG Department of Medicinal ChemistryAprather Weg 18A42096WuppertalGermany

The deletion of the C-terminal arginine of the anaphylatoxin protein C5a reduces it receptor binding *** how C-terminal arginine affects the structure and bioactivity of C5a is important for the development of C5a C-terminal mimics as drug ***,we report the total chemical synthesis of rat C5a and its D-enantiomer with its C-terminal arginine deleted,namely L-rC5a-desArg and *** structure of rC5a-desArg was then determined by racemic crystallography for the first *** C-terminal residues of rC5a-Arg were found to expand from the fourth helix in a continuous helical confo *** C-terminal conformation is significantly different from that of the previously reported full-length of C5a,indicating that the deletion of C-terminal arginine residue could result in the destruction of a positively charged surface formed by two adjacent Arg residues in C5a.

关键词： Anaphylatoxin C5a Racemic crystallization Solid phase peptide synthesis Chemical protein synthesis Hydrazide-based native chemical ligation

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

EFFECT OF Na,K AND Ce ON CRACK PROPAGATION RESISTANCE FOR ALLOY 8090

引用

Acta Metallurgica Sinica(English Letters) 1992年第5期5卷 336-340页

作者： MENG Liang zhang baochang LIANG Ying zhang Weizhong TIAN Puke Northwestern Polytechnical University,Xi’an,China Northwestern Polytechnical Univ Xi'an China

The crack propagation resistance for alloy 8090 sheet decreases with the increase of content of Na and K *** improvement over the fracture toughness and crack initiation resist- ance for the alloy containing Na and K within certain limit may be made by adding 0.05% *** beneficial modification would not be manifested if the alloy contained more Na and *** of the reasons why Na and K injure the toughness of the alloy might be due to that they cause the precipitation of T_1 phase along grain and subgrain boundaries and accelerate the abnormal growth of recystallized grains.

关键词： alloy 8090 crack propagation resistance Na K Ce

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

Crowd Counting for Static Images: A Survey of Methodology

Crowd Counting for Static Images: A Survey of Methodology

引用

第三十九届中国控制会议

作者： Ying Luo Jinhu Lu baochang zhang School of Automation Science and Electrical Engineering Beihang University Shenyuan Honors College Beihang University School of Automation Science and Electrical Engineering State Key Laboratory of Software Development Environmentand Beijing Advanced Innovation Center for Big Data and Brain ComputingBeihang University

Crowd counting for static images is one of the typical application fields of image processing. it is essential to direct crowd analysis for public security purposes under the background of a rapidly growing social environment. Crowd counting for static images involves crucial challenges such as occlusions, scale variations, scene perspective distortions and diverse crowd distributions. It has now become an independent research hotspot. This paper investigates the literature of crowd counting tasks, summarizes the evolution of counting approaches, compares and analyzes traditional methods and current trends of CNN estimation-based counting, and then evaluates common datasets for crowd counting tasks. Conclusions concerning summarizations and prospects are made.

关键词： Crowd counting crowd analysis pattern recognition computer vision

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

Deepwalk-aware graph convolutional networks

引用

Science China(Information Sciences) 2022年第5期65卷 81-95页

作者： Taisong JIN Huaqiang DAI Liujuan CAO baochang zhang Feiyue HUANG Yue GAO Rongrong JI Media Analytics and Computing Lab Department of Computer Science and Technology School of InformaticsXiamen University Media Analytics and Computing Lab Department of Artificial Intelligence School of InformaticsXiamen University School of Automation Science and Electrical Engineering Beihang University Tencent Youtu Lab School of Software Tsinghua University

Graph convolutional networks(GCNs) provide a promising way to extract the useful information from graph-structured data. Most of the existing GCNs methods usually focus on local neighborhood information based on specific convolution operations, and ignore the global structure of the input data. To extract the latent representation for the graph-structured data more effectively, we introduce a deepwalk strategy into GCNs to efficiently explore the global graph information. This strategy can complement the local neighborhood information of a graph, resulting in the more robust representation for the graph *** fusion of the local neighboring and global structured information of a graph can further facilitate deep feature learning at the output layer of GCNs for node classification. Experimental results show that the proposed model has achieved state-of-the-art results on three benchmark datasets including Cora, Citeseer,and Pubmed citation networks.

关键词： graph convolutional networks global information fusion node classification

来源：

同方期刊数据库

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 Noise Estimation Method Based on Envelope Pseudo-measurement System in Adaptive Kalman Filter

A Noise Estimation Method Based on Envelope Pseudo-measureme...

引用

第43届中国控制会议

作者： Liuyang Jiang Guohui Zheng baochang zhang School of Information and Control Engineering Qingdao University of Technology Hangzhou Innovation Institute Beihang University

For adaptive Kalman filter algorithm,the accurate estimation of the measurement noise covariance matrix(R) is particularly important. Recently,researchers have developed an enhanced algorithm that can adaptively estimate R using second-order mutual-difference(SOMD) sequences generated by redundant measurements. However,the applicability of this algorithm to most real-world systems is limited because it relies on redundant measurements. In order to introduce SOMD algorithm into non-redundant measurement system,the envelope pseudo-measurement noise covariance estimation(EPMNCE)is constructed in this paper. By extracting the upper and lower envelope of the signal,the median of each sampling point is calculated to form a pseudo-measurement system. On this basis,the feedback detection mechanism is introduced to judge the validity of the envelope,and the virtual detection and missing detection are corrected to ensure that the observed values are within the upper and lower envelope range. By comparing the simulation results of Extended Kalman filtering(EKF),Innovative-based adaptive estimation(IAE) and EPMNCE,it is found that the results of EPMNCE are smoother and closer to the actual value. The application of EPMNCE can improve the accuracy and stability of noise covariance estimation.

关键词： envelope pseudo-measurement noise covariance estimation adaptive Kalman filter

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

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：