Name
Abstract | ||
|---|---|---|
The cloud causes the invalid observation from sensor aboard on satellite on cloudy day, which will corrupt the spatio-temporal continuity of the land surface parameters retrieved based on remote sensing, and prevent the fusing multi-source remote sensing data in the field of the quantitative remote sensing. Based on the requirement of quantitative remote sensing, this paper proposes an improved method which considers the anisotropic reflectance of land surface and the spatial heterogeneity, and carried out the proposed method in Heihe river basin of China to retrieve MODIS band surface reflectance. the proposed method first constructs the spatial distribution based on the corresponding BRDF and solar-viewing geometry; then, a geographically weighted regression (GWR) was introduced into the proposed method to derive the MODIS band surface reflectance at cloudy pixels one by one. And, the validation for the proposed method shows an accuracy (all of RMSE is 0.011, 4.9%), which is rational and good based on the criterions of the image processing and the quantitative remoter sensing. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1109/IGARSS.2016.7730113 | 2016 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS) |
Keywords | Field | DocType |
Anisotropic Reflectance, BRDF, Cloudy Pixel, Spatial Heterogeneity, GWR | Bidirectional reflectance distribution function,Meteorology,Satellite,Anisotropy,Computer science,Remote sensing,Image processing,Mean squared error,Spatial heterogeneity,Pixel,Spatial distribution | Conference |
ISSN | Citations | PageRank |
2153-6996 | 0 | 0.34 |
References | Authors | |
4 | 5 | |