Pesquer L., Domingo C., Pons X. (2013) A geostatistical approach for selecting the highest quality MODIS daily images. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 7887 LNCS: 608-615.LinkDoi: 10.1007/978-3-642-38628-2_72
The aim of this work was to develop a new methodology for automatic selection of the highest quality MODIS daily images, MOD09GA Surface Reflectance product. The methodology developed here complements the quality assessment of MODIS products with a geostatistical analysis of spatial pattern images based on variogram tools. The resulting selection is formed by 26 high-quality images (from an initial dataset of 365) from throughout 2007. Most images with geometric distortion problems, such as the bow-tie effect, were rejected. The automatic selection was validated by comparing it to manual selection, which showed that it achieved an overall accuracy of 71.4%. © 2013 Springer-Verlag.
Subscribe to our Newsletter to get the lastest CREAF news.
© 2016 CREAF | Legal notice