Operational algorithm for ice–water classification on dual-polarized RADARSAT-2 images

Abstract. Synthetic Aperture Radar (SAR) data from RADARSAT-2 (RS2) in dual-polarization mode provide additional information for discriminating sea ice and open water compared to single-polarization data. We have developed an automatic algorithm based on dual-polarized RS2 SAR images to distinguish open water (rough and calm) and sea ice. Several technical issues inherent in RS2 data were solved in the pre-processing stage, including thermal noise reduction in HV polarization and correction of angular backscatter dependency in HH polarization. Texture features were explored and used in addition to supervised image classification based on the support vector machines (SVM) approach. The study was conducted in the ice-covered area between Greenland and Franz Josef Land. The algorithm has been trained using 24 RS2 scenes acquired in winter months in 2011 and 2012, and the results were validated against manually derived ice charts of the Norwegian Meteorological Institute. The algorithm was applied on a total of 2705 RS2 scenes obtained from 2013 to 2015, and the validation results showed that the average classification accuracy was 91 ± 4 %.

[1]  David A. Clausi,et al.  Automated Ice–Water Classification Using Dual Polarization SAR Satellite Imagery , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[2]  Torbjørn Eltoft,et al.  Comparison of feature based segmentation of full polarimetric SAR satellite sea ice images with manually drawn ice charts , 2013 .

[3]  Torbjørn Eltoft,et al.  Comparison of automatic segmentation of full polarimetric SAR sea ice images with manually drawn ice charts , 2013 .

[4]  Wolfgang Dierking,et al.  Sea Ice Monitoring by Synthetic Aperture Radar , 2013 .

[5]  Ola M. Johannessen,et al.  Classification of Sea Ice Types in ENVISAT Synthetic Aperture Radar Images , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[6]  Leif Toudal Pedersen,et al.  Monitoring sea ice using Envisat ASAR — A new era starting 10 years ago , 2012, 2012 IEEE International Geoscience and Remote Sensing Symposium.

[7]  J. Karvonen,et al.  Operational SAR-based sea ice drift monitoring over the Baltic Sea , 2012 .

[8]  David A. Clausi,et al.  Operational SAR Sea-Ice Image Classification , 2012, IEEE Transactions on Geoscience and Remote Sensing.

[9]  David A Clausi,et al.  Feature extraction of dual-pol SAR imagery for sea ice image segmentation , 2012 .

[10]  John J. Yackel,et al.  Evaluation of C-band SAR polarimetric parameters for discrimination of first-year sea ice types , 2012 .

[11]  Wolfgang Dierking,et al.  Mapping of Different Sea Ice Regimes Using Images From Sentinel-1 and ALOS Synthetic Aperture Radar , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[12]  David A Clausi,et al.  MAGIC: MAp-Guided Ice Classification System , 2010 .

[13]  Mohammed Shokr,et al.  Compilation of a radar backscatter database of sea ice types and open water using operational analysis of heterogeneous ice regimes , 2009 .

[14]  J. Yackel,et al.  Sea ice type and open water discrimination using dual co-polarized C-band SAR , 2009 .

[15]  David A. Clausi,et al.  IRGS: Image Segmentation Using Edge Penalties and Region Growing , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Ola M. Johannessen,et al.  Remote Sensing of Sea Ice in the Northern Sea Route: Studies and Applications , 2006 .

[17]  David A. Clausi,et al.  Operational map-guided classification of SAR sea ice imagery , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[18]  Marko Mäkynen,et al.  Open water detection from Baltic Sea ice Radarsat-1 SAR imagery , 2005, IEEE Geoscience and Remote Sensing Letters.

[19]  Ola M. Johannessen,et al.  Multisensor approach to automated classification of sea ice image data , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[20]  Leen-Kiat Soh,et al.  ARKTOS: an intelligent system for SAR sea ice image classification , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[21]  Rashpal S. Gill,et al.  SAR Ice Classification Using Fuzzy Screening Method , 2003 .

[22]  Martti Hallikainen,et al.  Incidence angle dependence of the statistical properties of C-band HH-polarization backscattering signatures of the Baltic Sea ice , 2002, IEEE Trans. Geosci. Remote. Sens..

[23]  T. Markus,et al.  ERS SAR characterization of coastal polynyas in the Arctic and comparison with SSM/I and numerical model investigations , 2002 .

[24]  David A Clausi.  An analysis of co-occurrence texture statistics as a function of grey level quantization , 2002 .

[25]  Leen-Kiat Soh,et al.  Texture analysis of SAR sea ice imagery using gray level co-occurrence matrices , 1999, IEEE Trans. Geosci. Remote. Sens..

[26]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[27]  M. Shokr.  Evaluation of second‐order texture parameters for sea ice classification from radar images , 1991 .

[28]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[29]  Karolin Baecker,et al.  Remote Sensing Of Sea Ice In The Northern Sea Route Studies And Applications , 2016 .

[30]  Stian Normann Anfinsen,et al.  Assessing polarimetric SAR sea-ice classifications using consecutive day images , 2015, Annals of Glaciology.

[31]  Jochen Horstmann,et al.  A Descalloping Postprocessor for ScanSAR Images of Ocean Scenes , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[32]  F. Albregtsen.  Statistical Texture Measures Computed from Gray Level Coocurrence Matrices , 2008 .

[33]  T. Vinje,et al.  The ice transport through the Fram Strait , 1986 .

[34]  Michael Unser,et al.  Sum and Difference Histograms for Texture Classification , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.