A NEW AUTOMATED-METHOD OF CLOUD MASKING FOR ADVANCED VERY HIGH-RESOLUTION RADIOMETER FULL-RESOLUTION DATA OVER THE OCEAN

This paper reports on a new algorithm to remove cloud-contaminated pix els from daytime and nighttime -km advanced very high resolution radio meter (AVHRR) data. The technique was developed in response to Navy ne eds to efficiently and accurately eliminate cloud contaminated pixels from real-time satellite digital images. The remaining ''cloud free'' sea surface temperature (SST) pixels would be available for analysts t o utilize in tracking ocean mesoscale fronts and eddies as well as inp ut to SST and ocean thermal analysis. Initial poor results with existi ng cloud-masking techniques led to this effort. The method uses a seri es of approaches to locate cloud-contaminated pixels which include (1) use of multiple bands to detect signatures not readily available from single-channel data, (2) extraction of cloud edge information through local segmentation of the image using the cluster shade texture measu re based on the gray level cooccurrence (GLC) matrix and, (3) discrimi nation of cloud-free from cloud-contaminated regions with an area labe ling procedure. This technique is evaluated on identical data sets uti lizing other experimental and operational cloud algorithms and cloud m asks produced through human interpretation. This method, tested over a wide range of conditions and geographical locations, produces accurat e and efficient daytime and nighttime SST data sets.