检索结果-南通市图书馆

COMPrompter: reconceptualized segment anything model with multiprompt network for camouflaged object detection

Science China(Information Sciences) 2025年第1期68卷 189-203页

作者： xiaoqin zhang Zhenni YU Li ZHAO Deng-Ping FAN Guobao XIAO Zhejiang Province Key Laboratory of Intelligent Informatics for Safety and Emergency Wenzhou University Nankai International Advanced Research Institute (SHENZHEN FUTIAN) College of Computer Science Nankai University School of Computer Science and Technology Tongji University

We rethink the segment anything model(SAM) and propose a novel multiprompt network called COMPrompter for camouflaged object detection(COD). SAM has zero-shot generalization ability beyond other models and can provide an ideal framework for COD. Our network aims to enhance the single prompt strategy in SAM to a multiprompt strategy. To achieve this, we propose an edge gradient extraction module, which generates a mask containing gradient information regarding the boundaries of camouflaged objects. This gradient mask is then used as a novel boundary prompt, enhancing the segmentation process. Thereafter, we design a box-boundary mutual guidance module, which fosters more precise and comprehensive feature extraction via mutual guidance between a boundary prompt and a box prompt. This collaboration enhances the model's ability to accurately detect camouflaged objects. Moreover, we employ the discrete wavelet transform to extract high-frequency features from image embeddings. The high-frequency features serve as a supplementary component to the multiprompt ***, our COMPrompter guides the network to achieve enhanced segmentation results, thereby advancing the development of SAM in terms of COD. Experimental results across COD benchmarks demonstrate that COMPrompter achieves a cutting-edge performance, surpassing the current leading model by an average positive metric of 2.2% in COD10K. In the specific application of COD, the experimental results in polyp segmentation show that our model is superior to top-tier methods as well. The code will be made available at https://***/guobaoxiao/COMPrompter.

关键词： segment anything model camouflaged object detection boundary prompt

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

Stochastic Approximation for Expensive One-Bit Feedback Systems

引用

Tsinghua Science and Technology 2017年第3期22卷 317-327页

作者： xiaoqin zhang Huimin Ma Jinghuan Wen Department of Electronic Engineering Tsinghua UniversityBeijing 100084China

One-bit feedback systems generate binary data as their output and the system performance is usually measured by the success rate with a fixed parameter combination. Traditional methods need many executions for parameter optimization. Hence, it is impractical to utilize these methods in Expensive One-Bit Feedback Systems （EOBFSs）, where a single system execution is costly in terms of time or money. In this paper, we propose a novel algorithm, named Iterative Regression and Optimization （IRO）, for parameter optimization and its corresponding scheme based on the Maximum Likelihood Estimation （MLE） method and Particle Swarm Optimization （PSO） method, named MLEPSO-IRO, for parameter optimization in EOBFSs. The IRO algorithm is an iterative algorithm, with each iteration comprising two parts： regression and optimization. Considering the structure of IRO and the Bernoulli distribution property of the output of EOBFSs, MLE and a modified PSO are selected to implement the regression and optimization sections, respectively, in MLEPSO-IRO. We also provide a theoretical analysis for the convergence of MLEPSO-IRO and provide numerical experiments on hypothesized EOBFSs and one real EOBFS in comparison to traditional methods. The results indicate that MLEPSO-IRO can provide a much better result with only a small amount of system executions.

关键词： stochastic approximation parameter optimization one-bit feedback system regression MaximumLikelihood Estimation (MLE)

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 the Occlusion Strategy in Visual Tracking

引用

Tsinghua Science and Technology 2016年第2期21卷 221-230页

作者： Jinghuan Wen Huimin Ma xiaoqin zhang Department of Electronic Engineering Tsinghua University

Interference and anti-interference are two opposite and important issues in visual tracking. Occlusion interference can disguise the features of a target and can also be used as an effective benchmark to determine whether a tracking algorithm is reliable. In this paper, we proposed an inner Particle Swarm Optimization （PSO） algorithm to locate the optimal occlusion strategy under different tracking conditions and to identify the most effective occlusion positions and direction of movement to allow a target to evade tracking. This algorithm improved the standard PSO process in three ways. First, it introduced a death process, which greatly reduced the time cost of optimization. Second, it used statistical data to determine the fitness value of the particles so that the fitness more accurately described the tracking. Third, the algorithm could avoid being trapped in local optima, as the fitness changes with time. Experimental results showed that this algorithm was able to identify a global optimal occlusion strategy that can disturb the tracking machine with 86.8% probability over more than 10 000 tracking processes. In addition, it reduced the time cost by approximately 80%, compared with conventional PSO algorithms.

关键词： particle swarm optimization virtual tracking occlusion interference death process statistical fitnessfunction

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

LIDAR:learning from imperfect demonstrations with advantage rectification

引用

Frontiers of Computer Science 2022年第1期16卷 57-66页

作者： xiaoqin zhang Huimin MA Xiong LUO Jian YUAN Department of EE Tsinghua UniversityBeijing100084China School of Computer and Communication Engineering University of Science and Technology BeijingBeijing100083China

In actor-critic reinforcement learning(RL)algorithms,function estimation errors are known to cause ineffective random exploration at the beginning of training,and lead to overestimated value estimates and suboptimal *** this paper,we address the problem by executing advantage rectification with imperfect demonstrations,thus reducing the function estimation *** with expert demonstrations has been widely adopted to accelerate the learning process of deep reinforcement learning when simulations are expensive to ***,existing methods,such as behavior cloning,often assume the demonstrations contain other information or labels with regard to performances,such as optimal assumption,which is usually incorrect and useless in the real *** this paper,we explicitly handle imperfect demonstrations within the actor-critic RL frameworks,and propose a new method called learning from imperfect demonstrations with advantage rectification(LIDAR).LIDAR utilizes a rectified loss function to merely learn from selective demonstrations,which is derived from a minimal assumption that the demonstrating policies have better performances than our current *** learns from contradictions caused by estimation errors,and in turn reduces estimation *** apply LIDAR to three popular actor-critic algorithms,DDPG,TD3 and SAC,and experiments show that our method can observably reduce the function estimation errors,effectively leverage demonstrations far from the optimal,and outperform state-of-the-art baselines consistently in all the scenarios.

关键词： learning from demonstrations actor-critic reinforcement learning advantage rectification

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

Constructing vertical phase separation of polymer blends via mixed solvents to enhance their photovoltaic performance

引用

Science China Chemistry 2015年第2期58卷 309-316页

作者： Qingqing Yang Jiantai Wang xiaoqin zhang Jun zhang Yingying Fu Zhiyuan Xie State Key Laboratory of Polymer Physics and Chemistry Changchun Institute of Applied ChemistryChinese Academy of Sciences University of Chinese Academy of Sciences

A polymer blend comprising poly(3-hexylthiophene)(P3HT)donor and poly[2,7-(9,9′-octyl-fluorene)-alt-5,5-(4′,7′-di-2-thienyl-5′,6′-bis(hexyloxy)-2′,1′,3′-benzothiadiazole)](PFDTBT-OC6)acceptor is used as the active layer to fabricate all-polymer solar *** blend morphology variance processed with pure and mixed solvents,and the related photovoltaic performance,are investigated in *** is found that,due to its low surface energy,a thin P3HT enrichment layer on the top surface of the active layer greatly increases bimolecular recombination and results in S-kinks of the illuminated current density-voltage *** the incorporation of p-xylene(a marginal solvent of P3HT)in the blend solution,the P3HT enrichment atop the active layer surface is effectively decreased because the high boiling-point p-xylene suppresses the diffusion of P3HT chains toward the top surface during the film-drying *** bimolecular recombination was thus improved and the S-kinks of the photovoltaic curves were completely *** overall power conversion efficiencies of the devices are strongly boosted(from 0.88%to 1.41%)when chlorobenzene:p-xylene mixed solvent is used to replace pure chlorobenzene.

关键词： polymer solar cells polymer blends vertical phase separation bimolecular recombination S-kink

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 vapor deposition of amorphous molybdenum sulphide on black phosphorus for photoelectrochemical water splitting

引用

Journal of Materials Science & Technology 2021年第9期68卷 1-7页

作者： Yuanmei Xu xiaoqin zhang Zhihong Chen Krzysztof Kempa Xin Wang Lingling Shui International Academy of Optoelectronics at Zhaoqing South China Normal UniversityGuangzhou510006China Guangdong Provincial Key Laboratory of Optical Information Materials and Technology South China Academy of Advanced OptoelectronicsSouth China Normal UniversityGuangzhou510006China State Key Lab of Silicon Materials Zhejiang UniversityHangzhou310027China Key Laboratory for Water Quality and Conservation of the Pearl River Delta Ministry of EducationInstitute of Environmental Research at Greater BayGuangzhou UniversityGuangzhou510006China Department of Physics Boston CollegeMA02467USA

Non-precious metal electrocatalyst molybdenum sulphide(MoS) and black phosphorus(BP) are highly promising catalysts for H_(2) evolution reaction(HER).However,BP is environmentally unstable and the basal planes of crystal MoS_(2) are inactive toward ***,amorphous molybdenum sulphide(MoSx)directly on BP/BiVO4 film dramatically improves the performance of photoelectrochemical water splitting compared with pure ***,we demonstrate that a BP layer,inserted between the MoSx and BiVO4,can enhance the photoelectrochemical performance and improve the stability of the ***,MoS_(x)/B P/BVO electrode shows the excellent current density of 2.1 mA/cm^(2) at the potential of 1.2 V(vs Ag/AgCl),which is twice higher than that of pure BVO *** novel nanostructure materials will lead to a new class of non-precious metal photocatalysts for hydrogen production.

关键词： Amorphous Molybdenum sulphide Black phosphorus Photoanode Water splitting

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

DcnnGrasp: towards accurate grasp pattern recognition with adaptive regularizer learning

引用

Science China(Information Sciences) 2024年第12期67卷 323-324页

作者： xiaoqin zhang Ziwei HUANG Jingjing ZHENG Shuo WANG Xianta JIANG College of Computer Science and Artificial Intelligence Wenzhou University Department of Computer Science Memorial University of Newfoundland

Vision-based grasp pattern recognition1)identifies the specific hand pose（fingers and palm configuration） for the object to be grasped based on visual information of household objects, which benefits tasks such as upper limb prosthesis control and gesture-based human-robot interaction. Various methods, including convolutional neural networks（CNN）,have been proposed for grasp pattern recognition, leading to advancements in fields such as robot imitation, humanrobot interaction, prosthetic design, and robot control.

关键词： Contrastive Learning

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

Modeling and simulation of an anatomy teaching system

引用

Visual Computing for Industry,Biomedicine,and Art 2019年第1期2卷 67-74页

作者： xiaoqin zhang Jingyi Yang Na Chen Shaoxiang zhang Yifa Xu Liwen Tan Institute of Digital Medicine School of Biomedical Engineering and Medical ImagingArmy Medical University(Third Military Medical University)Chongqing 400038China Shandong Digihuman Technology Co. IncJinan 250101China

Specimen observation and dissection have been regarded as the best approach to teach anatomy,but due to the severe lack of anatomical specimens in recent years,the quality of anatomy teaching has been seriously *** order to disseminate anatomical knowledge effectively under such circumstances,this study discusses three key factors(modeling,perception,and interaction)involved in constructing virtual anatomy teaching systems in *** ensure the authenticity,integrity,and accuracy of modeling,detailed three-dimensional(3D)digital anatomical models are constructed using multi-scale data,such as the Chinese Visible Human dataset,clinical imaging data,tissue sections,and other *** anatomical knowledge ontology is built according to the needs of the particular teaching *** kinds of anatomical knowledge and 3D digital anatomical models are organically combined to construct virtual anatomy teaching system by means of virtual reality equipment and *** perception of knowledge is realized by the Yi Chuang Digital Human Anatomy Teaching System that we have *** virtual interaction mode,which is similar to actual anatomical specimen observation and dissection,can enhance the transmissibility of anatomical *** virtual anatomy teaching system captures the three key *** can provide realistic and reusable teaching resources,expand the new medical education model,and effectively improve the quality of anatomy teaching.

关键词： Chinese visible human Anatomy knowledge ontology Virtual reality Anatomy teaching

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

Parameter estimation method based on parameter function surface

引用

Science China(Technological Sciences) 2013年第9期56卷 1485-1498页

作者： BAO WeiMin zhang xiaoqin ZHAO LiPing State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering Hohai University College of Hydrology and Water Resources Hohai University

By analyzing the structure of the objective function based on error sum of squares and the information provided by the objective function, the essential problems in the current parameter estimation methods are summarized: (1) the information extracted from the objective function based on error sum of squares is unreasonable or even wrong for parameter estimation; and (2) the surface of the objective function based on error sum of squares is more complex than that of the parameter function, which indicates that the optimal parameter values should be searched on the surface of the parameter function instead of the objective function. This paper proposes the concept of sample intersection and demonstrates the uniqueness theorem of intersection point (namely the uniqueness of optimal parameter values). According to the characteristics of parameter function surface and Taylor series expansion, a parameter estimation method based on the sample intersection information extracted from parameter function surface (PFS method) was constructed. The results of theoretical analysis and practical application show that the proposed PFS method can avoid the problems in the current automatic parameter calibration, and has fast convergence rate and good performance in parameter calibration.

关键词： parameter estimation objective function surface parameter function surface uniqueness theorem intersection Taylor series

来源：

维普期刊数据库

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

Early Activation of Astrocytes does not Affect Amyloid Plaque Load in an Animal Model of Alzheimer's Disease

引用

Neuroscience Bulletin 2018年第6期34卷 912-920页

作者： Dongpi Wang xiaoqin zhang Mingkai Wang Dongming Zhou Hongyu Pan Qiang Shu Binggui Sun Department of Neurobiology NHC and CAMS Key Laboratory of Medical Neurobiology Zhejiang University School of Medicine Department of Neurology of the First Affiliated Hospital NHC and CAMS Key Laboratory of Medical NeurobiologyZhejiang University School of Medicine Children's Hospital Zhejiang University School of Medicine

Astrocytes are closely associated with Alzheimer’s disease(AD). However, their precise roles in AD pathogenesis remain controversial. One of the reasons behind the different results reported by different groups might be that astrocytes were targeted at different stages of disease progression. In this study, by crossing h APP(human amyloid precursor protein)-J20 mice with a line of GFAP-TK mice, we found that astrocytes were activated specifically at an early stage of AD before the occurrence of amyloid plaques, while microglia were not affected by this crossing. Activation of astrocytes at the age of 3–5 months did not affect the proteolytic processing of hAPP and amyloid plaque loads in the brains of hAPP-J20 mice. Our data suggest that early activation of astrocytes does not affect the deposition of amyloid b in an animal model of AD.

关键词： Astrocyte Alzheimer's disease Amyloid plaque Mouse

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：