Comparison of GOES Cloud Classification Algorithms Employing Explicit and Implicit Physics

Abstract Cloud-type classification based on multispectral satellite imagery data has been widely researched and demonstrated to be useful for distinguishing a variety of classes using a wide range of methods. The research described here is a comparison of the classifier output from two very different algorithms applied to Geostationary Operational Environmental Satellite (GOES) data over the course of one year. The first algorithm employs spectral channel thresholding and additional physically based tests. The second algorithm was developed through a supervised learning method with characteristic features of expertly labeled image samples used as training data for a 1-nearest-neighbor classification. The latter’s ability to identify classes is also based in physics, but those relationships are embedded implicitly within the algorithm. A pixel-to-pixel comparison analysis was done for hourly daytime scenes within a region in the northeastern Pacific Ocean. Considerable agreement was found in this analysis,...

[1]  Kenneth Sassen,et al.  Cloud Type and Macrophysical Property Retrieval Using Multiple Remote Sensors , 2001 .

[2]  R. Welch,et al.  Automated Cloud Classification of Global AVHRR Data Using a Fuzzy Logic Approach , 1997 .

[3]  Bryan A. Baum,et al.  A Grouped Threshold Approach for Scene Identification in AVHRR Imagery , 1999 .

[4]  Michael J. Pavolonis,et al.  Global Daytime Distribution of Overlapping Cirrus Cloud from NOAA’s Advanced Very High Resolution Radiometer , 2005 .

[5]  Matthias Steiner,et al.  The Oceanic Convection Diagnosis and Nowcasting System , 2009 .

[6]  Johannes Schmetz,et al.  Monitoring deep convection and convective overshooting with METEOSAT , 1997 .

[7]  E. Williams,et al.  The Identification and Verification of Hazardous Convective Cells over Oceans Using Visible and Infrared Satellite Observations , 2008 .

[8]  Anne Fouilloux,et al.  Assessment of Clouds Characteristics from Satellite Observations by Means of Self-Organized Neural Networks , 1998 .

[9]  W. Paul Menzel,et al.  Comparison between current and future environmental satellite imagers on cloud classification using MODIS , 2007 .

[10]  Taneil Uttal,et al.  Daytime Global Cloud Typing from AVHRR and VIIRS: Algorithm Description, Validation, and Comparisons , 2005 .

[11]  Steven A. Ackerman,et al.  Cloud Detection with MODIS. Part II: Validation , 2008 .

[12]  Michael J. Pavolonis,et al.  Daytime Cloud Overlap Detection from AVHRR and VIIRS , 2004 .

[13]  Richard A. Frey,et al.  Automated cloud detection and classification of data collected by the Visible Infrared Imager Radiometer Suite (VIIRS) , 2005 .

[14]  Karl-Göran Karlsson,et al.  NWCSAF AVHRR Cloud Detection and Analysis Using Dynamic Thresholds and Radiative Transfer Modeling. Part II: Tuning and Validation , 2005 .

[15]  Richard L. Bankert,et al.  Optimization of an Instance-Based GOES Cloud Classification Algorithm , 2007 .

[16]  John S. DaPonte,et al.  Comparison of genetic algorithm systems with neural network and statistical techniques for analysis of cloud structures in midlatitude storm systems , 1997, Pattern Recognit. Lett..

[17]  David W. Aha,et al.  Improvement to a Neural Network Cloud Classifier , 1996 .

[18]  Paul M. Tag,et al.  An AVHRR Multiple Cloud-Type Classification Package , 2000 .

[19]  W. Paul Menzel,et al.  Operational retrieval of atmospheric temperature, moisture, and ozone from MODIS infrared radiances , 2003, SPIE Asia-Pacific Remote Sensing.

[20]  Michael J. Garay,et al.  An operational MISR pixel classifier using support vector machines , 2007 .

[21]  S. Lakshmivarahan,et al.  Development of an Automated Classification Procedure for Rainfall Systems , 2005 .

[22]  Andrew Heidinger,et al.  Rough Draft CLAVR-x Cloud Mask Algorithm Theoretical Basis Document , 2004 .