大区域TM影像归一化植被指数季相归一化处理方法

Seasonal Normalization Processing Methods of TM NDVI Images in Large Areas

针对高分辨率遥感监测植被时大范围获取影像的季相往往不一致,影响了归一化植被指数的连续性与可比性问题,借助MODIS归一化植被指数产品,开展TM归一化植被指数季相归一化方法研究。以福建省为例,分析不同季相MODIS归一化植被指数之间的一次项、二次项最小二乘关系及差值关系,对比分析转换结果。结果表明,二次项关系优于其他两种效果,较好地消除归一化植被指数时相不一致问题,获得了较为满意的效果。利用MODIS归一化植被指数构建二次项,完成福建省TM归一化植被指数季相归一化,消除了不同景影像间的季相不一致,使在大区域范围内归一化植被指数具有空间上的可比性和一致性。

Moderate resolution remote sensing data is widely used in vegetation monitoring.However,seasonal aspects of large areas is easily inconsistent in acquired images,which impair the continuity and comparability of normalized difference of vegetation index.MODIS data has the characteristics of high temporal resolution and wide coverage.The correlation analysis of TM NDVI and MODIS NDVI in different seasons shows that the variation of TM NDVI is quite similar to that of MODIS NDVI in the vegetation covered area.Therefore,with the MODIS NDVI products data,this paper studies TM NDVI on the seasonal normalization methods.By making a demonstration based on Fujian province,we analyze the relationship of quadratic least squares,the relationship of linear least squares and differential relationship between MODIS NDVI in different season phases.Subsequently,we analysis the converted results.The results show that the relationship of quadratic least squares method is superior to the other two methods,which can better eliminate the inconsistency of NDVI seasonal aspects and obtain satisfactory results.By using MODIS NDVI products,we build a quadratic term least squares relationship and fulfill the seasonal normalized of TM NDVI in Fujian province and eliminate the inconsistency of NDVI seasonal aspects at the same time,making the NDVI data in large areas has comparability and consistency on the space.