中文摘要：

采用了基于重归一化植被指数的像元二分模型，并结合实测数据和相关辅助数据，对三江源自然保护区2007年的植被覆盖度进行了估算和精度验证．植被覆盖度实测值和遥感估算值之间的线性相关系数达到了0．9017，平均精度为81．52％．结果表明使用此方法进行大面积植被覆盖度的遥感估算是有效可行的．

英文摘要：

A dimidiate pixel model was established based on re-normalized difference vegetation index. Using this model, the vegetation coverage of the Sanjiangyuan national nature reserve in 2007 was estimated under the support of field investigation data and related auxiliary data, which were further validated according to the field investigation data. It was found that the correlation coefficient between the field investigation data and the imagery extracted values was 0.901 7, and the mean precision was 81.52%. Our study confirms that this is a feasible approach in the estimation of vegetation coverage over large area from remote sensing data.