基于Ts-EVI时间序列谱的冬小麦面积提取

Extracting winter wheat area using temporal sequence of Ts-EVI

冬小麦是中国重要的粮食作物,利用遥感技术准确获取冬小麦分布面积对于从宏观上实现冬小麦生长状况和病虫害监测预警等田间精细管理以及产量预测具有十分重要的意义。采用NASA提供的2005年2～6月MYD09A1和MYD11A2数据产品,结合冬小麦生育期形态变化特征,以Ts和EVI构建的时间序列谱对河北省冬小麦种植面积的提取方法进行了研究。结果表明:冬小麦在Ts-EVI特征空间中具有独特的序列谱相特征,返青期后冬小麦的Ts在整体上表现为逐渐升高,EVI在开花期后下降,灌浆后期快速降低;对Ts-EVI空间信息的一维化处理后,采用主成分分析方法对时间信息降维,以ISODATA非监督分类方法实现冬小麦分离提取结果的样点验证精度为91.20%;Ts-EVI时间序列谱反映的作物生长过程的生理意义明确,采用Ts-EVI时间序列谱的遥感分类方法可以较好地实现冬小麦的遥感分类提取,具有较好的可行性。

Winter wheat is one of the most important crops in China. It is significant to extract winter wheat area accurately using remote sensing technology because area information can be used for monitoring growth condition and estimating yield. In this study, temporal series of MYD09A1 and MYD11A2 productions provided by NASA from February to June in 2005 were used to establish temporal sequence of Ts-EVI. Variation characteristics of winter wheatphenology were analyzed and the method to extract winter area using temporal sequence of Ts-EVI was also discussed. Results show that： （1） winter wheat has the particular temporal sequence characteristics in Ts-EVI space. Ts of winter wheat rises with the time after the greenup and EVI increases before heading and milking stages then begins to descend; （2） one dimensional treatment was done for Ts-EVI space information and principal component analysis was used to reduce the dimensions of temporal sequence images, then ISODATA, the unsupervised method was used to extract winter wheat area. The accuracy of classification of winter wheat tested with in situ samples was 91.20%; （3） it is definite to express the physiological meaning with temporal sequence of Ts-EVI. Remote sensed classifying method with temporal sequence of Ts-EVI can extract winter wheat successfully, which shows great feasibility.