检索结果-南通市图书馆

An improved deep forest model for forecast the outdoor atmospheric corrosion rate of low-alloy steels

在线全文

学校读者我要写书评

暂无评论

Journal of Materials Science & Technology 2020年第14期47卷 202-210页

作者： Yuanjie Zhi Tao Yang dongmei fu School of Automation and Electrical Engineering University of Science and Technology BeijingBeijing100083China Beijing Engineering Research Center of Industrial Spectrum Imaging University of Science and Technology BeijingBeijing 100083China

The paper proposes a new deep structure model,called Densely Connected Cascade Forest-Weighted K Nearest Neighbors(DCCF-WKNNs),to implement the corrosion data modelling and corrosion ***,we collect 409 outdoor atmospheric corrosion samples of low-alloy steels as experiment ***,we give the proposed methods process,including random forests-K nearest neighbors(RF-WKNNs)and ***,we use the collected datasets to verify the performance of the proposed *** results show that compared with commonly used and advanced machine-learning algorithms such as artificial neural network(ANN),support vector regression(SVR),random forests(RF),and cascade forests(cForest),the proposed method can obtain the best prediction *** addition,the method can predict the corrosion rates with variations of any one single environmental variable,like pH,temperature,relative humidity,SO2,rainfall or Cl-.By this way,the threshold of each variable,upon which the corrosion rate may have a large change,can be further obtained.

关键词： Random forests Deep forest model Low-alloy steels Outdoor atmospheric corrosion Prediction and data-mining

Modelling method with missing values based on clustering and support vector regression

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Engineering and Electronics 2010年第1期21卷 142-147页

作者： Ling Wang dongmei fu Qing Li Zhichun Mu Information Engineering School University of Science and Technology Beijing Beijing 100083 R R. China

Most real application processes belong to a complex nonlinear system with incomplete information. It is difficult to estimate a model by assuming that the data set is governed by a global model. Moreover, in real processes, the available data set is usually obtained with missing values. To overcome the shortcomings of global modeling and missing data values, a new modeling method is proposed. Firstly, an incomplete data set with missing values is partitioned into several clusters by a K-means with soft constraints （KSC） algorithm, which incorporates soft constraints to enable clustering with missing values. Then a local model based on each group is developed by using SVR algorithm, which adopts a missing value insensitive （MVI） kernel to investigate the missing value estimation problem. For each local model, its valid area is gotten as well. Simulation results prove the effectiveness of the current local model and the estimation algorithm.

关键词： modeling missing value K-means with soft constraints clustering missing value insensitive kernel.

Data mining to effect of key alloying elements on corrosion resistance of low alloy steels in Sanya seawater environmentAlloying Elements

在线全文

学校读者我要写书评

暂无评论

Journal of Materials Science & Technology 2021年第5期64卷 222-232页

作者： Xin Wei dongmei fu Mindong Chen Wei Wu Dequan Wu Chao Liu School of Automation and Electrical Engineering University of Science and Technology BeijingBeijing100083China Beijing Engineering Research Center of Industrial Spectrum Imaging University of Science and Technology BeijingBeijing 100083China SINOPEC Research Institute of Safety Engineering Qingdao266100China Beijing Advanced Innovation Center for Materials Genome Engineering Institute for Advanced Materials and TechnologyUniversity of Science and Technology BeijingBeijing100083China

In this paper,the relationship model between seawater environment,chemical composition and corrosion potential of low alloy steel is established and the distribution of corrosion potential of low alloy steel with changes in key alloying elements is *** research was carried out with the following steps:Firstly,the relationship model between corrosion potential of low alloy steel and its influencing factors was established by data dimension reduction and artificial neural network(ANN).Secondly,key alloying elements of experimental steels were selected out by Pearson correlation analysis,then the corrosion resistance element model was visualized to show the effect of key alloying elements on corrosion potential of low alloy ***,corrosion potential of low alloy steel with the change of key alloying elements was classified and visualized by classification *** mining results can reflect the validity of the proposed mining methods to a certain extent and provide an intuitive data basis for the development of high-quality and low-cost low alloy steels.

关键词： Low alloy steel Corrosion potential Key alloying elements Corrosion-resistant alloy Artificial neural network Data-driven model

Hand Target Extraction from Infrared Images with Descriptor based on Pixel Temporal Characteristics

在线全文

学校读者我要写书评

暂无评论

Hand Target Extraction from Infrared Images with Descriptor ...

基于人工靶向免疫疗法的红外手部痕迹目标提取

国防光电子论坛第二届新型探测技术及其应用研讨会

作者： dongmei fu Hui Li Tao Yang School of Automation and Electrical Engineering University of Science and Technology Beijing

Infrared trace image can record the trace left on the surface of objects after our body makes contacts with them and it is a special kind of blurred image with a transition zone. The traditional target extraction methods did not work well on trace image. This paper presents a target extraction method using an edge-shape descriptor based on pixel temporal characteristics to extract the original contour from sequence infrared images. Firstly, divide all the pixels of the image into three sets;then, use gray level values of a certain pixel in temporal sequence to describe the variation trend of the pixel and the Pearson correlation coefficient of two trends is used to represent the image feature of the point. Secondly, an edge-shape descriptor is used on the result of preliminary segmentation to judge which class a pixel belongs to. Finally, in further processing stage, use morphological operation to smooth the edge and remove isolated points. The experimental results show that the proposed method in this paper exhibits better extraction performance.

关键词： target extraction sequence infrared images pixel temporal characteristics edge-shape descriptor

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

电子与信息学报 2018年第2期40卷 346-352页

作者：付冬梅孙静杨焘北京科技大学自动化学院北京100083

红外手部热痕迹图像是特殊的模糊图像,该文提出一种人工靶向免疫疗法对其进行手部目标提取。首先依据序列图像中像素灰度的变化趋势设计了先天性免疫识别进行初分割;然后借鉴免疫的提呈机制,根据热扩散特性定义同心圆模板提取特征;基于... 详细信息

红外手部热痕迹图像是特殊的模糊图像,该文提出一种人工靶向免疫疗法对其进行手部目标提取。首先依据序列图像中像素灰度的变化趋势设计了先天性免疫识别进行初分割;然后借鉴免疫的提呈机制,根据热扩散特性定义同心圆模板提取特征;基于模板特征对模糊像素集适应性免疫识别;最后,指尖指谷病变检测分析,实施靶向治疗,保证了手的形态特征。与分水岭、SOM网络以及近几年研究成果进行了比较,表明提出的算法在目标提取率、绝对误差率均优于现有算法,提取结果更符合手的形态,同时扩展了应用热痕迹信息的时间跨度。

关键词：模糊红外图像模板特征免疫网络靶向免疫疗法

改进YOLOv3网络结构的遮挡行人检测算法

在线全文

学校读者我要写书评

暂无评论

模式识别与人工智能 2020年第6期33卷 568-574页

作者：刘丽郑洋付冬梅北京科技大学自动化学院北京100083 北京科技大学北京市工业波谱成像工程技术研究中心北京100083 北京科技大学顺德研究生院佛山528399

针对YOLOv3算法在监控视频行人检测中对遮挡目标漏检率较高的问题,文中提出改进YOLOv3网络结构的遮挡行人检测算法.首先在网络全连接层引入空间金字塔池化网络,增强网络的多尺度特征融合能力.然后采用网络剪枝的方式,精简网络冗余结构,... 详细信息

针对YOLOv3算法在监控视频行人检测中对遮挡目标漏检率较高的问题,文中提出改进YOLOv3网络结构的遮挡行人检测算法.首先在网络全连接层引入空间金字塔池化网络,增强网络的多尺度特征融合能力.然后采用网络剪枝的方式,精简网络冗余结构,避免网络层数加深导致的退化和过拟合问题,同时减少参数量.在走廊行人数据集上进行多尺度训练,获得最优的权重模型.实验表明,文中方法在平均准确率和检测速度上都有所提升.

关键词：行人检测深度学习 YOLOv3 空间金字塔池化网络网络剪枝

A Data-Driven Hybrid Machine Learning and Transfer Learning Algorithm for Long Consecutive Atmospheric Environment Missing Data Imputation

在线全文

学校读者我要写书评

暂无评论

A Data-Driven Hybrid Machine Learning and Transfer Learning ...

第43届中国控制会议

作者： Hao Meng Lizhen Shao dongmei fu Guangxuan Song Jintao Meng Beijing Engineering Research Center of Industrial Spectrum Imaging School of Automation and Electrical Engineering University of Science and Technology Beijing National Key Laboratory of Security Communication

Comprehensive and accurate environmental data form the foundation for studying the correlation between substance properties and the environment. Nevertheless, missing data is inevitable during the data collection process, posing a challenge to research advancement, especially in cases of prolonged and consecutive data gaps. The inherent periodicity of environmental data, which has rarely been considered separately in previous studies, plays an important role in imputing long consecutive missing data. This article combines the advantages of machine learning and transfer learning methods and proposes an imputation algorithm that can fully utilize the inherent periodic components of long consecutive missing data. The algorithm initially utilizes wavelet coherence to assess the periodic coherence of the data, decomposing it into periodic and non-periodic ***, the random forest method is employed to impute the periodic components, while the transfer learning approach is used to impute the non-periodic components. By aggregating the imputed values of each component, the final imputation output is obtained. The algorithm was validated using environmental data from five regions in China, demonstrating superior performance compared to existing algorithms across various levels of missing data.

关键词： Atmospheric environment Long consecutive missing data Data-driven modeling Periodic decomposition Transfer learning

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

时序增强的视频动作识别方法

模式识别与人工智能 2020年第10期33卷 951-958页

作者：张浩博付冬梅周珂北京科技大学自动化学院北京100083 北京科技大学顺德研究生院佛山528399 北京科技大学北京市工业波谱成像工程中心北京100083 北京科技大学高等工程师学院北京100083

针对视频动作识别中的时空建模问题,在深度学习框架下提出基于融合时空特征的时序增强动作识别方法.首先对输入视频应用稀疏时序采样策略,适应视频时长变化,降低视频级别时序建模成本.在识别阶段计算相邻特征图间的时序差异,以差异计算... 详细信息

针对视频动作识别中的时空建模问题,在深度学习框架下提出基于融合时空特征的时序增强动作识别方法.首先对输入视频应用稀疏时序采样策略,适应视频时长变化,降低视频级别时序建模成本.在识别阶段计算相邻特征图间的时序差异,以差异计算结果增强特征级别的运动信息.最后,利用残差结构与时序增强结构的组合方式提升网络整体时空建模能力.实验表明,文中算法在UCF101、HMDB51数据集上取得较高准确率,并在实际工业操作动作识别场景下,以较小的网络规模达到较优的识别效果.

关键词：动作识别深度学习时序增强结构工业监控视频

在线全文

学校读者我要写书评

暂无评论

中国共产党领导下的百年减贫实践及未来走向

财贸研究 2021年第8期32卷 1-8,56页

作者：张雄付冬梅上海财经大学人文学院上海200433 安徽财经大学马克思主义学院安徽蚌埠233030

经过百年实践,历经革命、建设与改革,中国共产党最终取得脱贫攻坚战的历史性胜利。始终坚持以人民为中心,以现代化道路为基础,充分发挥制度优势,探索形式多样的贫困治理模式是成功的关键。脱贫成绩固然令世人瞩目,但认清现代社会贫困背... 详细信息

经过百年实践,历经革命、建设与改革,中国共产党最终取得脱贫攻坚战的历史性胜利。始终坚持以人民为中心,以现代化道路为基础,充分发挥制度优势,探索形式多样的贫困治理模式是成功的关键。脱贫成绩固然令世人瞩目,但认清现代社会贫困背后的不同逻辑,更有助于对中国减贫实践进行正确定位。在普遍交往日益密切的现代世界,不同类型的贫困往往相互交织,只有自觉把握世界历史进程,认清中国社会贫困的逻辑,才能找准未来发展的方向。

关键词：世界历史进程减贫实践历史性贫困现代性贫困