...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Photogrammetric Engineering & Remote Sensing: Journal of the American Society of Photogrammetry >Mapping vegetation across large geographic areas: integration of remote sensing and GIS to classify multisource data
【24h】

Mapping vegetation across large geographic areas: integration of remote sensing and GIS to classify multisource data

机译：在较大的地理区域内绘制植被图：将遥感和GIS集成以对多源数据进行分类

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A two-stage classification process for mapping land cover across large geographic areas from digital imagery, such as Landsat TM products, is described. Stage 1 involves a two-pass, unsupervised classification designed to capture patterns evident ina color composite, followed by a merging of pixels into raster polygons based on spectral similarity. Stage 2 involves a supervised classification to label each raster polygon according to cover type (or other class feature). The second stage takes placein a GIS environment, after spectral and biophysical attributes are calculated for each polygon in the image. Classification accuracy is assessed using fuzzy sets, and individual GIS databases for adjacent images are virtually edge-matched, post-classification, to create seamless outputs across multiple scenes. We found that, by storing and analyzing data from separate scenes in separate databases, large geographic areas can be processed relatively quickly and efficiently.
机译：描述了用于从数字图像（例如Landsat TM产品）跨大地理区域映射土地覆盖的两阶段分类过程。第1阶段涉及两次通过的无监督分类，该分类旨在捕获彩色复合图像中明显的图案，然后基于光谱相似度将像素合并为光栅多边形。第2阶段涉及监督分类，以根据封面类型（或其他类别要素）标记每个栅格多边形。在为图像中的每个多边形计算光谱和生物物理属性之后，第二阶段在GIS环境中进行。使用模糊集评估分类准确性，对相邻图像的各个GIS数据库进行虚拟边缘匹配，后分类，以创建跨多个场景的无缝输出。我们发现，通过在不同的数据库中存储和分析来自不同场景的数据，可以相对快速有效地处理较大的地理区域。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号