期刊文献+

“三调”成果数据引导的耕地“非农化”遥感监测 被引量:11

Remote sensing monitoring for the non-agriculturalization of cultivated land guided by the third national land survey results data
原文传递
导出
摘要 为准确提取耕地“非农化”的变化信息,实现耕地“非农化”遥感监测,提出一种“三调”成果引导的耕地“非农化”遥感监测方法。先从前时相的矢量数据抽取矢量边界、类别属性等先验信息;进行超像素分割,提取像斑的光谱和纹理特征;在先验信息引导下,利用同质性特征分析、异常检测的方法自动筛选样本,利用影像特征差异分析的方法确定最优影像特征;针对不同的耕地变化类型,选择最优影像特征,结合支持向量机分类实现耕地“非农化”区域检测。在河北省选取两块区域进行实验,结果表明,该文方法在已有成果数据的引导下,能够有效检测出耕地“非农化”区域,总体检测精度均在97%以上。 In order to accurately extract the change information of the non-agriculturalization of cultivated land and realize the remote sensing monitoring of the non-agriculturalization of cultivated land,a remote sensing monitoring method is proposed for the non-agriculturalization of cultivated land guided by the third national land survey results data.Firstly,prior information,including vector boundaries and category attributes are extracted from the vector data in the previous phase.Secondly,super-pixel segmentation is used and spectral and texture features of image objects are extracted from high spatial resolution remote sensing images.Then,under the guidance of prior information,homogeneity feature analysis and outlier detection method are used to automatically screen samples,and image feature difference analysis is used to determine the optimal image features.Finally,for different change types of cultivated land,the optimal image features are selected,combined with support vector machine classification algorithm to achieve the non-agricultural area detection of cultivated land.Two experimental areas in Hebei Province are selected for experiments,Experiments results shows that can detect the non-agricultural areas of cultivated land effectively under the guidance of the existing results data,the overall detection accuracy both has reached more than 97%.
作者 李刚 宁晓刚 张翰超 王浩 郝铭辉 LI Gang;NING Xiaogang;ZHANG Hanchao;WANG Hao;HAO Minghui(Chinese Academy of Surveying and Mapping,Beijing 100036,China;Key Laboratory of Natural Resources Monitoring in Tropical and Subtropical area of South China,MNR,Guangzhou 510670,China)
出处 《测绘科学》 CSCD 北大核心 2022年第7期149-159,共11页 Science of Surveying and Mapping
基金 自然资源部华南热带亚热带自然资源监测重点实验室开放基金课题项目(2022NRMZ01) 国土空间规划专项监测与分析项目(A2113)
关键词 变化检测 成果数据引导 耕地“非农化” 光谱特征 纹理特征 支持向量机 change detection data guidance cultivated land conversion spectral feature texture feature support vector machine
作者简介 李刚(1996—),男,河南驻马店人,硕士研究生,主要研究方向为自然资源监测。E-mail:1390864129@qq.com;通信作者:宁晓刚,研究员,E-mail:ningxg@casm.ac.cn
  • 相关文献

参考文献15

二级参考文献146

共引文献332

同被引文献118

引证文献11

二级引证文献33

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部