作者: A. J. Schweiger , J. Key , J. A. Maslanik
DOI:
关键词:
摘要: A forward-feed back-propagation neural network is used to classify merged AVHRR and SMMR summer Arctic data. Four surface eight cloud classes are identified. Partial memberships of each pixel class examined for spectral ambiguities. Classification results compared manual interpretations those determined by a supervised maximum likelihood procedure. Results indicate that approach offers advantages in ease use, interpretability, utility indistinct time-variant classes.