Translator Disclaimer
29 August 2005 A new method for MODIS cloud classification
Author Affiliations +
A new technique called the local region of influence (LROI) scheme for supervised cloud classification of the Moderate Resolution Imaging Spectroradiometer (MODIS) is proposed. The classification of each observation is performed within the LROI, where the center of each class is calculated as a weighted average of its training class members with respect to each new observation. The probability of each class is assigned to each observation. The proposed LROI scheme is applied to the MODIS radiances observed from the scenes of clear skies, ice clouds, or water clouds. The classification results are compared with those from the maximum likelihood (ML) classification method, the multicategory support vector machine (MSVM) and the operational MODIS cloud mask algorithm. The lowest misclassification error rates show the advantage of the LROI scheme.
© (2005) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Bormin Huang, Steven A. Ackerman, and W. Paul Menzel "A new method for MODIS cloud classification", Proc. SPIE 5890, Atmospheric and Environmental Remote Sensing Data Processing and Utilization: Numerical Atmospheric Prediction and Environmental Monitoring, 58900X (29 August 2005);

Back to Top