Title :
Cloud classification using support vector machines
Author :
Azimi-Sadjadi, M.R. ; Zekavat, S.A.
Author_Institution :
Dept. of Electr. & Comput. Eng., Colorado State Univ., Fort Collins, CO, USA
Abstract :
Cloud classification from GOES 8 (Geostationary Operational Environmental Satellite) imagery data is performed using the infrared (IR) channel only. For each block of the image, first and second order statistics are extracted and used to train and test a classifier. In this paper, cloud classification is performed using a support vector machine (SVM) classifier. This scheme, which is typically used to solve two-class problems, has been extended to classify ten different cloud and no-cloud areas. Preliminary results indicate the promise of this method for meteorological applications
Keywords :
atmospheric techniques; clouds; geophysical signal processing; geophysics computing; image classification; GOES 8; atmosphere; classifier; cloud; cloud classification; computing; image classification; image processing; infrared; measurement technique; meteorology; optical imaging; satellite remote sensing; support vector machine; Brightness; Clouds; Data mining; Feature extraction; Meteorology; Neural networks; Risk management; Satellites; Support vector machine classification; Support vector machines;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2000. Proceedings. IGARSS 2000. IEEE 2000 International
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-6359-0
DOI :
10.1109/IGARSS.2000.861666
