论文部分内容阅读
光学和微波协同遥感反演对于提高农田土壤水分遥感反演精度十分重要。本文采用SMEX02数据集,研究了L波段土壤发射率与地表土壤水分之间的关系,分析了地面植被覆盖对L波段土壤发射率与地表水分之关系的影响规律,推导了以L波段土壤发射率和归一化植被指数NDVI为自变量的土壤水分反演模型。研究表明:L波段土壤发射率与地表土壤水分之间的相关性随NDVI的增加而下降。验证结果表明,本文算法相对常规经验算法,土壤水分反演精度明显提高,H极化条件下,土壤水分的反演精度RMSE由0.0553提高到0.0407,相关系数R2由0.70提高到0.81;V极化条件下,反演精度RMSE由0.0452提高到0.0348,相关系数R2由0.79提高到0.86。
Optical and microwave collaborative remote sensing inversion is very important for improving remote sensing accuracy of farmland soil moisture retrieval. In this paper, using the SMEX02 dataset, the relationship between L-band soil emissivity and surface soil moisture was studied. The influence of ground vegetation coverage on the relationship between L-band soil emissivity and surface water content was analyzed. The L-band soil emissivity And normalized vegetation index NDVI as independent variables of soil moisture inversion model. The results show that the correlation between L-band soil emissivity and surface soil moisture decreases with the increase of NDVI. The verification results show that the accuracy of soil moisture retrieval algorithm is significantly improved compared with the conventional empirical algorithm. Under H polarization, the RMSE of soil moisture is increased from 0.0553 to 0.0407, and the correlation coefficient R2 is increased from 0.70 to 0.81. V polarization Under the conditions, the RMSE of inversion accuracy increased from 0.0452 to 0.0348, and the correlation coefficient R2 increased from 0.79 to 0.86.