...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>International journal of remote sensing >Bracken fern frond status classification in the Andes of southern Ecuador: combining multispectral satellite data and field spectroscopy
【24h】

Bracken fern frond status classification in the Andes of southern Ecuador: combining multispectral satellite data and field spectroscopy

机译：厄瓜多尔南部安第斯山脉的蕨蕨叶状体状态分类：结合多光谱卫星数据和现场光谱

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

In the anthropogenic fire-disturbed ecosystem of the San Francisco Valley in the Andes of southeastern Ecuador, dense stands of an aggressive invasive weed, the southern bracken fern (Pteridium arachnoideum and Pteridium caudatum), dominate the landscape. To secure sustainable land management in the region, a comprehensive understanding of bracken spatial-distribution patterns and life cycle dynamics is crucial. We investigated the possibility of detecting bracken-infested areas and frond status (live, fungi-infected, and dead) by means of a high-resolution QuickBird scene from October 2010 and spectral signatures based on field spectroscopy. After image pre-processing, a two-step classification procedure first delineates the bracken-infested area by means of a maximum-likelihood hard classification. The probability-guided unmixing classifier with field-derived end-members is applied in the second step to obtain the fractional cover of the different frond statuses per pixel. The results showed that the areas infested by bracken could be distinguished from the other land-cover classes with high accuracy (overall accuracy of 0.9973). Also, the three frond statuses could be accurately classified at the sub-pixel level. The 'dead' class was the dominant frond status at the time of image acquisition (October 2010). We conclude that the extreme dry spell in October 2010 was particularly responsible for this dominance.
机译：在厄瓜多尔东南部安第斯山脉的旧金山山谷，人为的受火扰动的生态系统中，茂盛的侵略性杂草丛生，即南部蕨类蕨类植物（蕨菜蕨和尾状蕨菜）占主导地位。为了确保该地区的可持续土地管理，全面了解蕨菜的空间分布模式和生命周期动态至关重要。我们调查了通过2010年10月以来的高分辨率QuickBird场景和基于现场光谱学的光谱特征检测蕨菜感染区域和叶状体状态（活着，真菌感染和死亡）的可能性。在图像预处理之后，首先通过两步分类程序通过最大似然硬分类来确定蕨菜感染区域。在第二步骤中应用具有场派生的末端成员的概率指导的解混合分类器，以获得每个像素不同叶状状态的分数覆盖率。结果表明，蕨菜出没的地区可以与其他土地覆被类别区分开来，具有很高的准确度（总体准确度为0.9973）。同样，可以在子像素级别上对这三种叶状状态进行准确分类。在获取图像时（2010年10月），“死”类是主要的叶状体。我们得出的结论是，2010年10月的极端干旱是造成这种优势的主要原因。

著录项

  • 来源
    《International journal of remote sensing》 |2013年第20期|7020-7037|共18页
  • 作者

    G.F. Curatola Fernandez; B. Silva; J. Gawlik; B. Thies; J. Bendix;

  • 作者单位

    Laboratory for Climatology and Remote Sensing (LCRS), Faculty of Geography, University of Marburg, Marburg, Germany;

    Laboratory for Climatology and Remote Sensing (LCRS), Faculty of Geography, University of Marburg, Marburg, Germany;

    Institute of Geography, University of Erlangen-Nuernberg, Erlangen, Germany;

    Laboratory for Climatology and Remote Sensing (LCRS), Faculty of Geography, University of Marburg, Marburg, Germany;

    Laboratory for Climatology and Remote Sensing (LCRS), Faculty of Geography, University of Marburg, Marburg, Germany;

  • 收录信息 美国《科学引文索引》(SCI);美国《工程索引》(EI);
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

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