Minimum band set to classify AVHRR images

Abstract

Evaluates if multispectral IR satellite images can be classified using only two bands and how far from optimal such classification can be. Two different unsupervised classification algorithms for image classification have been applied to AHVRR (Advanced Very High Resolution Radiometer) images coming from the NOAA-14 (National Oceans and Atmospheric Administration) polar orbiter satellite. The goal is to assign each pixel to one of the following classes: sea, land, cloud, cloud edge or out of the scene. The last class is needed since the images are resampled with always the same geographical limit, thus, sometimes, due to the satellite being low on the horizon, some pixels can be out of the radiometer view. The first method is based on thresholds on the value of the NDVI (Normalized Differential Vegetation Index) alone. The second method is based on thresholds on 4 of the 5 bands of the AVHRR.