Analysis of variograms with various sample sizes from a multispectral image
Academic Article
Overview
Identity
Additional Document Info
Other
View All
Overview
abstract
Variogram plays a crucial role in remote sensing application and geostatistics. It is very important to estimate variogram reliably from sufficient data. In this study, the analysis of variograms computed on various sample sizes of remotely sensed data was conducted. A 100100 - pixel subset was chosen randomly from an aerial multispectral image which contains three wavebands, Green, Red and near-infrared (NIR). Green, Red, NIR and Normalized Difference Vegetation Index (NDVI) datasets were imported into R software for spatial analysis. Variograms of these four full image datasets and sub-samples with simple random sampling method were investigated. In this case, half size of the subset image data was enough to reliably estimate the variograms for NIR and Red wavebands. To map the variation on NDVI within the weed field, ground sampling interval should be smaller than 12 m. The information will be particularly important for Kriging and also give a good guide of field sampling on the weed field in the future study.
