Multilayer Perceptron Neural Networks Model for Meteosat Second Generation SEVIRI Daytime Cloud Masking

A multilayer perceptron neural network cloud mask for Meteosat Second Generation SEVIRI (Spinning Enhanced Visible and Infrared Imager) images is introduced and evaluated. The model is trained for cloud detection on MSG SEVIRI daytime data. It consists of a multi-layer perceptron with one hidden sigmoid layer, trained with the error back-propagation algorithm. The model is fed by six bands of MSG data (0.6, 0.8, 1.6, 3.9, 6.2 and 10.8 μm) with 10 hidden nodes. The multiple-layer perceptrons lead to a cloud detection accuracy of 88.96%, when trained to map two predefined values that classify cloud and clear sky. The network was further evaluated using sixty MSG images taken at different dates. The network detected not only bright thick clouds but also thin or less bright clouds. The analysis demonstrated the feasibility of using machine learning models of cloud detection in MSG SEVIRI imagery.

[1]  Michel Desbois,et al.  Automatic classification of clouds on Meteosat imagery - Application to high-level clouds , 1982 .

[2]  Shaun Quegan,et al.  Cloud Masking for Remotely Sensed Data Using Spectral and Principal Components Analysis , 2012 .

[3]  R. Saunders,et al.  An improved method for detecting clear sky and cloudy radiances from AVHRR data , 1988 .

[4]  Thomas H. Vonder Haar,et al.  A Bispectral Method for Cloud Parameter Determination , 1977 .

[5]  Jyoti Deshmukh,et al.  Relative radiometric correction of cloudy multitemporal satellite imagery , 2009, ICAC3 '09.

[6]  W. Rossow,et al.  The International Satellite Cloud Climatology Project (ISCCP): The First Project of the World Climate Research Programme , 1983 .

[7]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[8]  Lucas Alados-Arboledas,et al.  Using a Sky Imager for aerosol characterization , 2008 .

[9]  R. R. Rhinehart,et al.  A method to determine the required number of neural-network training repetitions , 1999, IEEE Trans. Neural Networks.

[10]  Evaluation of the Daylight Cycle of Model-Predicted Cloud Amount and Condensed Water Path over Europe with Observations from MSG SEVIRI , 2009 .

[11]  M. Derrien,et al.  Improvement of cloud detection near sunrise and sunset by temporal-differencing and region-growing techniques with real-time SEVIRI , 2010 .

[12]  Francisco Javier López Aligué,et al.  A Comparative Study of Two Neural Models for Cloud Screening of Iberian Peninsula Meteosat Images , 2001, IWANN.

[13]  Sibbo van der Veen,et al.  Cloud Detection Using Meteosat Imagery and Numerical Weather Prediction Model Data , 2000 .

[14]  Richard L. Bankert,et al.  Cloud Classification of AVHRR Imagery in Maritime Regions Using a Probabilistic Neural Network , 1994 .

[15]  S. Kox,et al.  Retrieval of cirrus cloud optical thickness and top altitude from geostationary remote sensing , 2014 .

[16]  Enio Bueno Pereira,et al.  A Simple Method for the Assessment of the Cloud Cover State in High-Latitude Regions by a Ground-Based Digital Camera , 2006 .

[17]  Warren G. Julian,et al.  Sky analysis from CCD images: cloud cover , 2001 .

[18]  Soteris A. Kalogirou,et al.  Artificial intelligence techniques for photovoltaic applications: A review , 2008 .

[19]  Yu Liu,et al.  An Improved Cloud Classification Algorithm for China’s FY-2C Multi-Channel Images Using Artificial Neural Network , 2009, Sensors.

[20]  Michel Desbois,et al.  Cloud Cover Analysis from Satellite Imagery Using Spatial and Temporal Characteristics of the Data , 1987 .

[21]  Vincenzo Cuomo,et al.  A Technique for Classifying Uncertain MOD35/MYD35 Pixels Through Meteosat Second Generation-Spinning Enhanced Visible and Infrared Imager Observations , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[22]  U. Bhosle,et al.  Relative Radiometric Correction of Cloudy Multitemporal Satellite Imagery , 2009 .

[23]  M. Derrien,et al.  Automatic cloud detection applied to NOAA-11 /AVHRR imagery , 1993 .

[24]  William J. Emery,et al.  An Automated Neural Network Cloud Classifier for Use over Land and Ocean Surfaces , 1997 .

[25]  Hartwig Deneke,et al.  A threshold-based cloud mask for the high-resolution visible channel of Meteosat Second Generation SEVIRI , 2013 .

[26]  R. Roebeling,et al.  An automated day-time cloud detection technique applied to MSG-SEVIRI data over Western Europe , 2010 .

[27]  Olivier Jourdan,et al.  Mesoscale cloud pattern classification over ocean with a neural network using a new index of cloud variability , 2006 .

[28]  F. Bretherton,et al.  Cloud cover from high-resolution scanner data - Detecting and allowing for partially filled fields of view , 1982 .

[29]  Gordon B. Davis,et al.  Automatic Estimation of Cloud Amount Using Computer Vision , 1992 .

[30]  J. Schmetz,et al.  AN INTRODUCTION TO METEOSAT SECOND GENERATION (MSG) , 2002 .

[31]  K. Karlsson,et al.  On the optimal method for evaluating cloud products from passive satellite imagery using CALIPSO-CALIOP data: example investigating the CM SAF CLARA-A1 dataset , 2013 .

[32]  Alfred J Prata,et al.  Observations of volcanic ash clouds in the 10-12 μm window using AVHRR/2 data , 1989 .

[33]  M. Derrien,et al.  MSG/SEVIRI cloud mask and type from SAFNWC , 2005 .

[34]  Scott T. Acton,et al.  Cloud tracking by scale space classification , 2002, IEEE Trans. Geosci. Remote. Sens..

[35]  M. Ahn,et al.  An exploratory study of cloud remote sensing capabilities of the Communication, Ocean and Meteorological Satellite (COMS) imagery , 2007 .

[36]  Ronald M. Welch,et al.  A neural network approach to cloud classification , 1990 .