遥感影像在采集过程中,地面覆盖种类数量庞大且采集影像清晰度低、分辨率较差,关键像素特征之间的阈值衡量标准模糊,导致信息提取难度增大,从而降低信息利用率。为此,提出了基于像素紧密程度的多源遥感影像信息提取方法。利用Contourle...遥感影像在采集过程中,地面覆盖种类数量庞大且采集影像清晰度低、分辨率较差,关键像素特征之间的阈值衡量标准模糊,导致信息提取难度增大,从而降低信息利用率。为此,提出了基于像素紧密程度的多源遥感影像信息提取方法。利用Contourlet变换,实现遥感影像空间域、变换域的多角度增强,优化遥感影像整体清晰度。利用简单线性迭代聚类(Simple Linear Iterative Clustering,SLIC)超像素算法计算像素聚类中心与邻近像素紧密程度,摆脱固定阈值影响。引入灰度共生矩阵(Gray-level Co-occurrenceMatrix,GLCM),提取主体特征信息;构建相关向量机分类模型,结合拉普拉斯二次逼近回归算法将提取问题转化为噪声回归问题,并对其展开求解,进而实现遥感影像的信息提取。实验结果表明:所提方法对遥感信息主体的分类与真实遥感信息主体分类基本一致,在信息提取过程中的错提取率和漏提取率低,总体提取精度保持在99%以上,且对道路信息提取清晰度高,表明该方法提高了遥感信息的解译水平。展开更多
Nowadays remote laboratories suffer the absence of reusability.In addition,their construction and maintenance require time,money and skills.The system implementation of a specific remote lab is neither generic nor reu...Nowadays remote laboratories suffer the absence of reusability.In addition,their construction and maintenance require time,money and skills.The system implementation of a specific remote lab is neither generic nor reusable.In this paper,a solution for a reusable remote lab dedicated for disparate types of scientific and engineering experiments is presented. The experiment designer needs only to connect the experiment components and equipment such as capacitors,resistors, transistors,function generators with a switch system of a lab server,then,she/he has to map this connection structure in a configuration data structure.Once a student starts the Web-based client user-interface and logs-in into the lab server, the menu structure of the graphical user-interface builds and initializes itself automatically,using information stored in a configuration data structure.This contribution discusses some hitherto used lab servers,some of their drawbacks,the desirable requirements on a universal remote lab,which simplify the building process of newer lab experiments consisting of experiment components and equipment as well as a client user-interface that could enable students to remotely access the experiment.展开更多
文摘遥感影像在采集过程中,地面覆盖种类数量庞大且采集影像清晰度低、分辨率较差,关键像素特征之间的阈值衡量标准模糊,导致信息提取难度增大,从而降低信息利用率。为此,提出了基于像素紧密程度的多源遥感影像信息提取方法。利用Contourlet变换,实现遥感影像空间域、变换域的多角度增强,优化遥感影像整体清晰度。利用简单线性迭代聚类(Simple Linear Iterative Clustering,SLIC)超像素算法计算像素聚类中心与邻近像素紧密程度,摆脱固定阈值影响。引入灰度共生矩阵(Gray-level Co-occurrenceMatrix,GLCM),提取主体特征信息;构建相关向量机分类模型,结合拉普拉斯二次逼近回归算法将提取问题转化为噪声回归问题,并对其展开求解,进而实现遥感影像的信息提取。实验结果表明:所提方法对遥感信息主体的分类与真实遥感信息主体分类基本一致,在信息提取过程中的错提取率和漏提取率低,总体提取精度保持在99%以上,且对道路信息提取清晰度高,表明该方法提高了遥感信息的解译水平。
文摘Nowadays remote laboratories suffer the absence of reusability.In addition,their construction and maintenance require time,money and skills.The system implementation of a specific remote lab is neither generic nor reusable.In this paper,a solution for a reusable remote lab dedicated for disparate types of scientific and engineering experiments is presented. The experiment designer needs only to connect the experiment components and equipment such as capacitors,resistors, transistors,function generators with a switch system of a lab server,then,she/he has to map this connection structure in a configuration data structure.Once a student starts the Web-based client user-interface and logs-in into the lab server, the menu structure of the graphical user-interface builds and initializes itself automatically,using information stored in a configuration data structure.This contribution discusses some hitherto used lab servers,some of their drawbacks,the desirable requirements on a universal remote lab,which simplify the building process of newer lab experiments consisting of experiment components and equipment as well as a client user-interface that could enable students to remotely access the experiment.
