检索结果-南通市图书馆

Ghost-YOLO v8: An Attention-Guided Enhanced Small Target Detection Algorithm for Floating Litter on Water Surfaces

Computers, Materials and Continua 2024年第3期80卷 3713-3731页

作者： huangfu z. Li S. Yan L. School of Information Engineering North China University of Water Resources and Electric Power Zhengzhou 450046 China

Addressing the challenges in detecting surface floating litter in artificial lakes, including complex environments, uneven illumination, and susceptibility to noise and weather, this paper proposes an efficient and lightweight Ghost-YOLO (You Only Look Once) v8 algorithm. The algorithm integrates advanced attention mechanisms and a small-target detection head to significantly enhance detection performance and efficiency. Firstly, an SE (Squeez.-and-Excitation) mechanism is incorporated into the backbone network to fortify the extraction of resilient features and precise target localiz.tion. This mechanism models feature channel dependencies, enabling adaptive adjustment of channel importance, thereby improving recognition of floating litter targets. Secondly, a 160 × 160 small-target detection layer is designed in the feature fusion neck to mitigate semantic information loss due to varying target scales. This design enhances the fusion of deep and shallow semantic information, improving small target feature representation and enabling better capture and identification of tiny floating litter. Thirdly, to balance performance and efficiency, the GhostConv module replaces part of the conventional convolutions in the feature fusion neck. Additionally, a novel C2fGhost (CSPDarknet53 to 2-Stage Feature Pyramid Networks Ghost) module is introduced to further reduce network parameters. Lastly, to address the challenge of occlusion, a new loss function, WIoU (Wise Intersection over Union) v3 incorporating a flexible and non-monotonic concentration approach, is adopted to improve detection rates for surface floating litter. The outcomes of the experiments demonstrate that the Ghost-YOLO v8 model proposed in this paper performs well in the dataset Marine, significantly enhances precision and recall by 3.3 and 7.6 percentage points, respectively, in contrast with the base model, mAP@0.5 and mAP 0.5:0.95 improve by 5.3 and 4.4 percentage points and reduces the computational volu

关键词： attention mechanism ghostnet loss function small target detection head surface floating litter target detection YOLO v8

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

Study on Quantitative Precipitation Estimation by Polarimetric Radar Using Deep Learning

引用

Advances in Atmospheric Sciences 2024年第6期41卷 1147-1160页

作者： Jiang huangfu z.iqun HU Jiafeng z.ENG Lirong WANG Yongjie z.U State Key Lab of Severe Weather Chinese Academy of Meteorological SciencesBeijing 100081China Meteorological observatory of Jiangxi Province Nanchang 330096China Hebei Meteorological Disaster Prevention and Environmental Meteorological Center Shijiazhuang 050021China School of Atmospheric Sciences Chengdu University of Information TechnologyChengdu 610225China

Accurate radar quantitative precipitation estimation(QPE)plays an essential role in disaster prevention and *** this paper,two deep learning-based QPE networks including a single-parameter network and a multi-parameter network are ***,a self-defined loss function(SLF)is proposed during *** dataset includes Shijiaz.uang S-band dual polarimetric radar(CINRAD/SAD)data and rain gauge data within the radar’s 100-km detection range during the flood season of 2021 in North *** that the specific propagation phase shift(KDP)has a roughly linear relationship with the precipitation intensity,KDP is set to 0.5°km^(-1 )as a threshold value to divide all the rain data(AR)into a heavy rain(HR)and light rain(LR)***,12 deep learning-based QPE models are trained according to the input radar parameters,the precipitation datasets,and whether an SLF was adopted,*** results suggest that the effects of QPE after distinguishing rainfall intensity are better than those without distinguishing,and the effects of using SLF are better than those that used MSE as a loss function.A z.R relationship and a z.-KDP-R synthesis method are compared with deep learning-based *** mean relative errors(MRE)of AR models using SLF are improved by 61.90%,51.21%,and 56.34%compared with the z.R relational method,and by 38.63%,42.55%,and 47.49%compared with the synthesis ***,the models are further evaluated in three precipitation processes,which manifest that the deep learning-based models have significant advantages over the traditional empirical formula methods.

关键词： polarimetric radar quantitative precipitation estimation deep learning single-parameter network multi-parameter network

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

含钛渣制备多孔沸石分子筛吸附剂及其CO_(2)吸附性能

引用

精细化工 2024年第8期41卷 1726-1736,1856页

作者：皇甫林尚波罗中秋祖运周新涛钒钛资源综合利用国家重点实验室四川攀枝花617000 昆明理工大学化学工程学院云南昆明650500

以钒钛磁铁矿高炉冶炼产生的水淬渣为主要原料,经“酸处理-碱熔-水热晶化”制备了多孔沸石分子筛吸附剂(简称吸附剂,下同),采用XRD、SEM、FTIR对吸附剂进行了表征,考察了NaAlO_(2)添加量对吸附剂结构的影响,采用动态吸附法评价了吸附剂... 详细信息

以钒钛磁铁矿高炉冶炼产生的水淬渣为主要原料,经“酸处理-碱熔-水热晶化”制备了多孔沸石分子筛吸附剂(简称吸附剂,下同),采用XRD、SEM、FTIR对吸附剂进行了表征,考察了NaAlO_(2)添加量对吸附剂结构的影响,采用动态吸附法评价了吸附剂的CO_(2)吸附性能,考察了预处理温度、吸附温度和CO_(2)流量对吸附剂的CO_(2)吸附性能的影响,并利用动力学模型进行了数据拟合,分析了其吸附行为。结果表明,随着NaAlO_(2)添加量的增加,水热晶化产物依次完成了“无定形→初步结晶→FAU型→LTA型”沸石分子筛的结构转变;制备的吸附剂(z.4)经350℃预处理,在吸附温度25℃、CO_(2)流量分别为1.4 mL/min(固定床反应装置)和50 mL/min(同步热分析仪)的条件下,CO_(2)吸附效果最佳,其CO_(2)穿透与饱和吸附量为2.16、3.39 mmol/g;其CO_(2)吸附行为符合拟二级动力学模型,吸附过程由物理和化学吸附共同主导,拟二级动力学模型吸附速率常数为0.0785mmol/(g·s);经5次循环吸附/再生后,其CO_(2)饱和吸附量保持率为96.8%。

关键词：含钛渣水热合成多孔沸石分子筛 CO_(2)吸附吸附动力学功能材料

来源：

维普期刊数据库

同方期刊数据库博看期刊评论

在线全文

学校读者我要写书评

暂无评论

Role of Nitrite in Tumor Growth, Symbiogenetic Evolution of Cancer Cells, and China’s Successes in the War against Cancer

引用

Journal of Cancer Therapy 2011年第5期2卷 601-606页

作者： Kenneth J. Hsu Chao S. huangfu Min z. Qin 不详

Statistics and experiments indicate a correlation between cancer mortality and nitrite in drinking water. Nitrite is a reductant that can deprive a cell of oxygen;it is also an oxidant that can be a substrate in anaerobic ammonium oxidation, the metabolic mode of the anammox bacteria. Eukaryote cells evolved through a fusion of anaerobic and aerobic bacteria. We postulate that an anammox bacterium sought refuge in a fusion with a membrane-bound cluster of aerobic bacteria. While the latter evolved into mitochondria organelles, the former became the nucleus of a prokaryote cell. Eventually, oxidative phosphorilation is the characteristic metabolic pathway of normal eukaryote cells, and we postulate that anammox is the protein-catabolism pathway for cancer cells. The metabolism consumes nitrite and explains thus the link between nitrite and cancer.

关键词： Nitrite Tomor Symbiogenetic Evolution

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

Dynamics of unstable continental subduction:Insights from numerical modeling

引用

Science China Earth Sciences 2017年第2期60卷 218-234页

作者： huangfu PengPeng WANG YueJun FAN WeiMing LI z.ongHai z.OU Yongz.i School of Earth Science and Geological Engineering Sun Yat-sen University Key Laboratory of Computational Geodynamics College of Earth SciencesUniversity of Chinese Academy of Sciences CAS Center for Excellence in Tibetan Plateau Earth Sciences Key Laboratory of Continental Collision and Plateau UpliftInstitute of Tibetan Plateau ResearchChinese Academy of Sciences

Numerical experiments are used in this study to systematically investigate the effects of convergence rate,crustal rheological strength,and lithospheric thermal structure on the dynamics of continental *** study focuses on the types,conditions and processes of unstable continental *** results suggest that the development of unstable continental subduction can be promoted by conditions that tend to decrease rheological strength of the lithosphere,such as low crustal rheological strength,"hot"thermal structure of the lithosphere,or low convergence *** subduction mode can be further categoriz.d into three types:(1)multi-stage slab breakoff,(2)continuously"flowing"of fluid-like slab into the upper mantle,and(3)large-scale detachment of the thickened orogenic *** three types of unstable continental subduction are respectively associated with(1)a low convergence rate,(2)"hot"thermal structure of the lithosphere with a high convergence rate,and(3)moderate-high crustal rheological strength with a low convergence *** is also revealed that the evolution of crustal melting is dominated by the deformation pattern of continental collision,which is mainly controlled by crustal rheological *** modelling results have important implications for understanding of continental subduction mode selection under specific geodynamic conditions.

关键词： Unstable continental subduction Stable continental subduction Slab breakoff Detachment Partial melting

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

On the designing principles and optimiz.tion approaches of bio-inspired self-organiz.d network: a survey

引用

Science China(Information Sciences) 2013年第7期56卷 5-32页

作者： z.ANG z.ongShan huangfu Wei LONG KePing z.ANG Xu LIU XiaoYuan z.ONG Bin Institute of Advanced Network Technology and New Services (ANTS) University of Science and Technology Beijing (USTB) Beijing Engineering and Technology Center for Convergence Networks and Ubiquitous Services University of Science and Technology Beijing (USTB)

A plethora of studies on self-organiz.tion has been carried out in broad areas including chemistry, biology, astronomy, medical science, telecommunications, etc., in both academia and industry. Following the studies on swarm intelligence observed in social species, the artificial self-organiz.d systems are expected to exhibit some intelligent features (e.g., flexibility, robustness, decentraliz.d control, self-evolution, etc.) that may have made social species so successful in the biosphere. In this paper, the application of swarm intelligence in communications networks will be studied, and we survey different aspects of bio-inspired mechanisms and examine various algorithms that have been proposed to improve the performance of artificial systems. Some fundamental self-organiz.d networking (SON) mechanisms, designing principles and optimiz.tion approaches for artificial systems will then be investigated, followed by some well-known bio-inspired algorithms (e.g., cooperation, division of labor, distributed network synchroniz.tion, load balancing, etc.) as well as their applications to the maintenance/operation/optimiz.tion of artificial systems being analyz.d. Besides, some new emerging technologies, such as the Self-X capabilities and cognitive machine-to-machine (M2M) optimiz.tion for the 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE)/LTE-Advanced systems, are also surveyed. Finally, the remaining challenges to be faced in designing the future heterogeneous systems will be discussed.

关键词： self-organiz.d networking heterogeneous cognitive radio bio-inspired swarm intelligence decentraliz.d load balancing cooperation

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

门控心肌灌注显像左心室形态指数和偏心指数在评估心肌梗死患者左心室重构中的价值

引用

中华核医学与分子影像杂志 2024年第1期44卷 6-11页

作者：席小依王璐霞姚玘皇甫世豪肖钰昕武志芳武萍李莉闫蕊王跃涛杨敏福李思进山西医科大学第一医院核医学科、分子影像精准诊疗省部共建协同创新中心太原030001 常州市第一人民医院核医学科常州213003 首都医科大学附属北京朝阳医院核医学科北京100020

目的探讨左心室形态指数(SI)和偏心指数(EI)对左心室重构的评估价值。方法回顾性分析2016年1月至2020年9月间于山西医科大学第一医院行门控心肌灌注显像(GMPI)且确诊心肌梗死(MI)的患者324例[男264例、女60例, 年龄(62.5±11.8)岁]和健... 详细信息

目的探讨左心室形态指数(SI)和偏心指数(EI)对左心室重构的评估价值。方法回顾性分析2016年1月至2020年9月间于山西医科大学第一医院行门控心肌灌注显像(GMPI)且确诊心肌梗死(MI)的患者324例[男264例、女60例, 年龄(62.5±11.8)岁]和健康对照(HC)者113例[男47例、女66例, 年龄(57.8±10.7)岁]。通过定量门控SPECT (QGS)软件获得SI[舒张末期SI(EDSI)、收缩末期SI(ESSI)]、EI、左心室功能参数[舒张末期容积(EDV)、收缩末期容积(ESV)、左心室射血分数(LVEF)、室壁运动总评分(SMS)、室壁增厚总评分(STS)、高峰射血率(PER)、高峰充盈率(PFR)]。采用倾向性评分(PS)逆处理概率加权法(IPTW)平衡组间协变量。分析平衡后的MI组与HC组间SI、EI和左心室功能参数的差异及相关性。应用ROC曲线评价2组EDV、EDSI、ESSI、EI单独及联合评估左心室收缩功能受损的价值。采用两独立样本t检验、Pearson相关及Spearman秩相关分析、Delong检验等分析数据。结果 IPTW后MI组(319例)的EDSI、ESSI大于HC组(133例;EDSI:0.66±0.09和0.60±0.06;ESSI:0.59±0.11和0.47±0.07;t值:8.05、14.67, 均Pz.:-2.64~-2.18, P值:0.008~0.029), 在HC组(AUC:0.911)高于EDV、EDV-EDSI和EDV-(1-EI)(z.:-2.60~-2.43, P值:0.009~0.015)。结论 MI患者SI增大、EI减小提示左心室球形度增加、重构程度加重。SI、EI对评估左心室形态、预测左心室重构和左心室收缩功能受损有一定的临床应用价值。

关键词：心肌梗死心室重构心肌灌注显像 99m锝甲氧基异丁基异腈

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

基于图索引过滤机制的三维CAD模型局部检索

引用

计算机集成制造系统 2015年第7期21卷 1679-1689页

作者：皇甫中民张树生西北工业大学现代设计与集成制造技术教育部重点实验室陕西西安710072

针对现有三维CAD模型局部检索方法效率低、难以满足工程实际需要的问题,提出一种基于图索引过滤机制的三维CAD模型局部检索方法。依据模型的B-Rep表示形式,提取其必要的几何拓扑属性等信息,将三维CAD模型用属性邻接标记图表示;依据图的... 详细信息

针对现有三维CAD模型局部检索方法效率低、难以满足工程实际需要的问题,提出一种基于图索引过滤机制的三维CAD模型局部检索方法。依据模型的B-Rep表示形式,提取其必要的几何拓扑属性等信息,将三维CAD模型用属性邻接标记图表示;依据图的谱理论为属性邻接标记图及其各节点的局部结构进行编码,提出引理以论证模型局部匹配需满足的必要条件,并据此提出一种两层过滤的策略,构建了一种图编码索引结构;模型局部检索采用过滤—验证的思想,先依据提出的索引和过滤策略剪枝不符合规则的模型,再对候选集中的模型采用子图匹配的方法进行验证并输出。实验结果表明,所提方法在保证检索质量的基础上能够显著提高局部检索的效率,可满足工程检索的实际需要。

关键词：三维CAD模型局部检索图谱理论索引过滤

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

基于层次特征描述子的三维CAD模型检索

引用

计算机集成制造系统 2015年第12期21卷 3095-3106页

作者：皇甫中民张树生闫雒恒华北水利水电大学信息工程学院河南郑州450046 西北工业大学现代设计与集成制造技术教育部重点实验室陕西西安710072

为提高三维CAD模型检索对模型局部细节特征的区别能力,提高相似性比较和检索效率,提出一种基于图谱及空间词袋表征的CAD模型层次特征描述子构建和检索方法。依据模型的B-Rep表示形式,提取其必要的几何拓扑属性等信息,将三维CAD模型用属... 详细信息

为提高三维CAD模型检索对模型局部细节特征的区别能力,提高相似性比较和检索效率,提出一种基于图谱及空间词袋表征的CAD模型层次特征描述子构建和检索方法。依据模型的B-Rep表示形式,提取其必要的几何拓扑属性等信息,将三维CAD模型用属性邻接标记图表示;将模型训练集合中的各模型分割成面面互连且具有一定工程意义的局部区域,依据图谱理论对其进行局部特征描述,并通过聚类构建词汇本;基于局部区域图谱及空间词袋表征构建CAD模型由细粒度到粗粒度的层次化特征描述子,并据此采取一种粗检索与精匹配相结合的两层检索机制进行三维CAD模型检索。实验结果表明,该方法能较好地区别模型局部细节特征,检索质量和效率均有明显提高,可有效支持CAD模型的重用。

关键词：三维CAD模型检索局部特征图谱空间词袋层次特征描述子

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

鱼群启发的三维CAD模型聚类与检索

引用

计算机辅助设计与图形学学报 2016年第8期28卷 1373-1382,1392页

作者：皇甫中民张树生闫雒恒华北水利水电大学信息工程学院郑州450046 西北工业大学现代设计与集成制造技术教育部重点实验室西安710072

为了提高三维CAD模型检索中模型的局部细节区别能力以及检索效率,提出一种鱼群启发的三维CAD模型聚类及检索方法.依据B-Rep形式CAD模型的属性邻接图以及图谱理论,采用一种融合空间邻接关系的词袋模式作为模型的特征描述子,用于模型聚类... 详细信息

为了提高三维CAD模型检索中模型的局部细节区别能力以及检索效率,提出一种鱼群启发的三维CAD模型聚类及检索方法.依据B-Rep形式CAD模型的属性邻接图以及图谱理论,采用一种融合空间邻接关系的词袋模式作为模型的特征描述子,用于模型聚类与检索中的特征描述及相似性计算;针对模型聚类问题,受鱼群运动模式启发,提出基于全局公告信息引导及模糊c均值修正的人工鱼群聚类算法,将模型库空间聚类划分为若干子空间;模型检索采取两层检索机制:首先通过隶属度函数将索引模型定位至相应搜索子空间,然后在较小的子空间内进行相似性比较.实验结果表明,该方法的特征描述子能较好地区别模型局部细节特征,模型库聚类效果较好,检索质量和效率均有明显提高,可有效地支持CAD模型的重用.

关键词：三维CAD模型聚类检索空间词袋图谱理论鱼群

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：