摘要:
MODIS归一化差值植被指数( MODIS-normalized difference vegetation index,MODIS-NDVI)时间序列产品能够连续反映植被的覆盖情况,是农作物遥感测量的重要数据源。为研究基于MODIS数据的油菜种植分布信息提取技术,选取湖北省为研究区,利用2008—2013年75个时相的MODIS-NDVI时序数据,结合农作物物候和地面调查样本等辅助资料,通过建立油菜种植面积提取模型,采用多次阈值比较方法提取了2009—2013年湖北省油菜种植分布信息,与统计数据比较,总体提取精度为85%左右。最后利用环境小卫星HJ-1A CCD数据进行精度验证,证明了MODIS-NDVI时序数据及本文方法在油菜种植面积提取中的可靠性,对掌握油菜种植面积和产量信息、加强农业生产管理、调整农业结构及辅助政府有关部门制定科学合理的农业政策具有重要意义。
Abstract:
MODIS vegetation index time series of products can reflect the coverage of vegetation continuously, and is an important data source for remote sensing of crops. The authors selected Hubei Province as the study area. Using the MODIS-normalized difference vegetation index ( MODIS-NDVI ) data of 75 phases in the period from 2008 to 2013, and combining the crop phenology with samples of ground survey and other auxiliary details, the authors established the model for extracting the area of rape growing, and then extracted the distribution of rape growing in Hubei Province in the period from 2009 to 2013 . The overall accuracy of the extraction was 85%, as shown by the verification of the HJ -1A CCD data of small environmental satellite, and this indicates that the reliability of the data of MODIS vegetation index time series and the study methods proposed in this paper for the extraction of the area of rape seed cultivation. The results achieved by the authors are very important for detecting the area of rape growing and information of output, reinforcing the management of agricultural production, adjusting the structure of agriculture, and assisting the government to formulate the agricultural policy scientifically and ra-tionally.
下载: