• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Asian conference on remote sensing;ACRS >A PRELIMINARY STUDY ON USING NDVI AND NDWI TIME SERIES FOR MONITORING ABANDONED FARMLANDS IN MOUNTAIN AREAS
【24h】

A PRELIMINARY STUDY ON USING NDVI AND NDWI TIME SERIES FOR MONITORING ABANDONED FARMLANDS IN MOUNTAIN AREAS

机译：利用NDVI和NDWI时间序列监测山地废弃农田的初步研究。

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This study preliminary evaluated the feasibility of Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI) time series for mapping abandoned farmlands in mountainous areas in Chiba prefecture, Japan. NDVI and NDWI time series was derived from 8-day composite of MODIS Surface-Reflectance Products (MOD09Q1 and MOD09A1) for the period from 2002 to 2013. Noise components in both time series data, which were induced mainly by cloud contamination and atmospheric variability, were reduced with the method based on Savitzky-Golay filter. The refined time series data were then decomposed into trend, seasonal, and reminder components. A simple linear regression model was fitted to the trend component for each pixel, and model parameters (i.e., intercept and slope) were estimated to be used as the features useful to detect abandoned farmlands in the pixel area. This idea was based on the assumption that the cessation of cropping activity and flood control in paddy would be reflected in the trend component of NDVI and NDWI time series, and its intercept and slope would have different pattern between farmlands and abandoned farmlands. Discrimination performance of the features was evaluated with the area under the receiver operating characteristic (ROC) curve (AUC). The results showed that all candidate features were poor (AUC ＜ 0.6) in terms of classification performance, being contrary to our intention. Considering the poor performance could be mainly caused by low spatial resolution of MOD09 products, we concluded that the use of remotely-sensed data with higher spatial resolution would be required for mapping abandoned farmlands in mountainous areas.
机译：这项研究初步评估了归一化植被指数（NDVI）和归一化水分指数（NDWI）时间序列用于绘制日本千叶县山区废弃农田的可行性。 NDVI和NDWI时间序列是从2002年至2013年的8天MODIS表面反射产品（MOD09Q1和MOD09A1）合成的。两个时间序列数据中的噪声成分主要是由云污染和大气多变性引起的，使用基于Savitzky-Golay滤波器的方法进行了减少。然后，将经过精炼的时间序列数据分解为趋势，季节和提醒组成部分。将简单的线性回归模型拟合到每个像素的趋势分量，并将模型参数（即截距和斜率）估计为可用于检测像素区域内废弃农田的特征。这个想法是基于这样的假设，即水稻活动的停止和防洪将反映在NDVI和NDWI时间序列的趋势分量上，并且其截距和坡度在农田和废弃农田之间将具有不同的模式。使用接收器工作特性（ROC）曲线（AUC）下的面积评估特征的识别性能。结果表明，所有候选特征在分类性能方面均较差（AUC ＜0.6），这与我们的意图背道而驰。考虑到性能不佳可能主要是由于MOD09产品的空间分辨率低引起的，我们得出的结论是，需要使用具有较高空间分辨率的遥感数据来绘制山区废弃农田的地图。

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