作物NDVI遥感信息中由于不同土壤背景的混入干扰，导致利用NDVI对作物长势监测及其生态参量预测的有效性降低，进一步研究复杂土壤背景对作物NDVI影响去除方法技术仍然是当今重要的研究课题。项目以黄淮海平原冬小麦农田为研究区，依据GF1/2/4//6卫星遥感不同空间尺度NDVI信息特征，预先研究播种期冬小麦农田土壤背景遥感信息及其时空变化特征并作为先验知识，基于光谱混合理论模型，研究土壤类型、土壤水分、秸秆残留对各生长期小麦NDVI信息影响敏感性权重及其关联性，研究不同品种小麦生长特性及其物候特征对植被覆盖度影响，提出快速有效的小麦NDVI信息复杂土壤背景去除理论方法，建立小麦NDVI土壤背景去除前后转换技术模式及转换系数查找表，解决包括历史存档的小麦NDVI土壤背景影响去除难题，推动利用多光谱遥感NDVI信息进行大空间区域、长时间序列的国内外作物长势监测和粮食估产等领域的深入有效应用。

The phenomenon that crop NDVI information at remote-sensing’s pixel scale is often mixed with different soil backgrounds, leads to the decrease of effectiveness of crop condition monitoring and its ecological parameter prediction by using the NDVI information. Study the influence and removal methods of soil backgrounds on crop NDVI information is still the principal research topics at present. This project uses winter wheat farmland of China's huang-huai-hai plain as the research area and studies the remote sensing’s information and its space-time variation characteristics, as prior knowledge, of soil background of wheat farmland in seeding time based on the different spatial scale of NDVI information characteristics of GF1/2/4/6 satellite. Based on the theoretical model of mixed spectrum, the influential sensitivity weight and correlation of soil types ,soil water content and straw residue to wheat NDVI of each growing periods is analyzed. The project studies the influence of growing and phenological characteristics of different wheat cultivars on vegetation coverage, proposes the effective theoretical methods and technology of removing soil background, creates the information conversion’s implementation patterns and lookup table of transformation matrix of before and after the soil background removal of wheat NDVI information, settles the remaining puzzles like soil background removal of wheat NDVI information, and deeply drives the effective use of multi-spectral NDVI information to monitor crop condition and estimate grain yield in large space region and long time series at home and abroad.