In this study,the multi-resolution Kalman filter(MKF)algorithm,which can handle multi-resolution problems with high computational efficiency,was used to blend two emissivity products:the Global LAnd Surface Satellite(...In this study,the multi-resolution Kalman filter(MKF)algorithm,which can handle multi-resolution problems with high computational efficiency,was used to blend two emissivity products:the Global LAnd Surface Satellite(GLASS)(BBE)product and the Advanced Spaceborne Thermal Emission and Reflection Radiometer(ASTER)narrowband emissivity(NBE)product.The ASTER NBE product was first converted into a BBE product.A new detrending method was used to transfer the BBEs into a process suitable for the MKF.The new detrending method was superior to the two existing methods.Finally,both the de-trended GLASS and ASTER BBE products were incorporated into the MKF framework to obtain the optimal estimation at each scale.Field measurements collected in North America were used to validate the integrated BBEs.Visually,the fusion map showed good continuity,with the exception of the border areas,and the quality of the fusion map was better than that of the original maps.The validation results indicate that the MKF improved the BBE product accuracy at the coarse scale.In addition,the MKF was capable of recovering missing pixels at a finer scale.展开更多
基金supported by the National Natural Science Foundation of China[grant number 41371323]the National High Technology Research and Development Program of China[grant number 2013AA121201].
文摘In this study,the multi-resolution Kalman filter(MKF)algorithm,which can handle multi-resolution problems with high computational efficiency,was used to blend two emissivity products:the Global LAnd Surface Satellite(GLASS)(BBE)product and the Advanced Spaceborne Thermal Emission and Reflection Radiometer(ASTER)narrowband emissivity(NBE)product.The ASTER NBE product was first converted into a BBE product.A new detrending method was used to transfer the BBEs into a process suitable for the MKF.The new detrending method was superior to the two existing methods.Finally,both the de-trended GLASS and ASTER BBE products were incorporated into the MKF framework to obtain the optimal estimation at each scale.Field measurements collected in North America were used to validate the integrated BBEs.Visually,the fusion map showed good continuity,with the exception of the border areas,and the quality of the fusion map was better than that of the original maps.The validation results indicate that the MKF improved the BBE product accuracy at the coarse scale.In addition,the MKF was capable of recovering missing pixels at a finer scale.
