检索结果-南通市图书馆

Hyperspectral Image Classification Based on Capsule Network

Chinese Journal of Electronics 2022年第1期31卷 146-154页

作者： MA Qiaoyu ZHANG Xin ZHANG Chunlei ZHOU Heng School of Science China University of Geosciences (Beijing) School of Statistics Beijing Normal University beijing zhongdi runde petroleum technology co. Ltd.

The co.ventional co.volutional neural network performs not well enough in the ground objects classification because of its insufficient ability in maintaining sensitive spectral information and characterizing the co.ariance of spatial structure, resulting from the narrow sensitive frequency band and co.plex spatial structure with diversity of hyperspectral remote sensing data which caused more serious phenomena of "same material, different spectra" and "different material, same spectra".Therefore, an improved capsule network is proposed and introduced into hyperspectral image target reco.nition. A co.volution structure co.bining shallow features and multi-scale depth features is put forward to reduce the phenomena of "different material, same spectra" firstly,and then the diversity of the spatial structure is expressed by the capsule vector and sub-capsule division in channel wise, so that the averaging effect of the co.volution process is weakened in the spectral domain and the spatial domain to reduce the phenomena of "same material, different spectra". By co.paring the experimental results on the hyperspectral data sets such as Indian Pines,Salinas, Tea Tree and Xiongan, the capsule network shows strong spatial structure expression ability, flexible deep and shallow feature fusion ability in multi-scale, and its accuracy in target reco.nition is better than that of co.ventional co.volutional neural networks, so it is suitable for the reco.nition of co.plex targets in hyperspectral images.

关键词： Capsule network Space-spectrum fusion Hyperspectral Classification Deep learning

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

Predicting Initial Formation Temperature for Deep Well Engineering with a New Method

引用

Journal of Earth Science 2015年第1期26卷 108-115页

作者： Fuzong Zhou Yucheng Xiong Ming Tian Department of Eco.omics and Management Tongren University Geotechnical Survey Department China Petroleum Engineering Co.LTD.Beijing Company technology Department Zhongdi North China (Beijing) Technological Engineering Co.LTD.

With the progress of science and technology, human beings explore the energy underground with thousands of meters. As a thermophysical parameter, initial formation temperature （IFT） plays an essential role in deep well engineering. However, it is not easy to predict the IFT accurately before drilling. This work uses a new method to analyze the effect factors of the underground temperature field, and assumes an artificial surface to eliminate the disturbance of the human errors and equipment errors on the surface temperature and thermal co.ductivity. co.sidering different distributions of the formation thermal co.ductivity and the rock radiogenic heat production, an optimized model was established. With this model, the paper predicted the bottom temperature of the main hole of the Chinese co.tinental Scientific Drilling （CCSD） as 132.80 ℃ at 4 725 m depth with 0.5% error. When the thermal co.duction is dominant in the formation, this simple method can predict the IFT distribution effectively for deep well in the exploration stage. However, it is almost impossible to avoid aquifers in the formation of drilling deep well, an existing drillhole including groundwater is needed to predict for testing the model＇s accuracy.

关键词： initial formation temperature deep well thermal co.ductivity radiogenic heat production.

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

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

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：