Verification of a Multispectral Cloud Classifier

Jukka Iivarinen, Markus Peura, Ari Visa

Helsinki University of Technology
Laboratory of Computer and Information Science
Rakentajanaukio 2 C, FIN-02150 Espoo
Finland

Abstract

The multispectral cloud classification scheme is implemented to automate the interpretation of satellite images. The developed classifier is capable of classifying satellite images taken round the year, during day and night. The classifier is fully automatic, and it can be adapted to changing situations with new examples. The classification is done in two phases, the clouds are detected and they are classified. There is a need for numerical and objective methods in evaluating the classifier. The automatic evaluation scheme to check the quality of the classifier is based on the comparison between the SYNOP observations and the satellite observations. Comparisons with other published results show that the classifier is working relatively well. The improvement of wintertime classification is the main objective for future work.
In 9th Scandinavian Conference on Image Analysis , Uppsala, Sweden, June 6-9, 1995.
Click here to download the paper in postscript format (186697 bytes in compressed format).
Back to Cloud Project Home Page