Cloud Classification of AVHRR Imagery in Maritime Regions Using a Probabilistic Neural Network

摘要: Abstract Using Advanced Very High Resolution Radiometer data, 16 pixel × sample areas are classified into one of ten output classes using a probabilistic neural network (PNN). The cirrus, cirrocumulus, cirrostratus, altostratus, nimbostratus, stratocumulus, stratus, cumulus, cumulonimbus, and clear. Over 200 features drawn from spectral, textural, physical measures computed the data for each area. input patterns presented to subset these selected by routine that indicates discriminatory potential feature. training testing used PNN obtained 95 expertly labeled images taken seven maritime regions; provide 1633 areas. Theoretical accuracy classifier is determined two methods. In hold-one-cut method, trained on all samples minus tested the, remaining sample. this technique, 79.8% the...