Improved superpixel-based polarimetric synthetic aperture radar image classification integrating color features

Abstract. Various polarimetric features including scattering matrix, covariance matrix, polarimetric decomposition results, and textural or spatial information have already been used for polarimetric synthetic aperture radar (PolSAR) image classification. However, color features are rarely involved. We propose an improved superpixel-based PolSAR image classification integrating color features. First, we extract the color information using polarimetric decomposition. Second, by combining the color and spatial information of pixels, modified simple linear iterative clustering is used to generate small regions called superpixels. Then we apply Wishart distance to the superpixels to classify them into different classes. This method is demonstrated using the L-band Flevoland PolSAR data from AirSAR and Oberpfaffenhofen PolSAR data from ESAR. The results show that this method works well for areas with homogeneous terrains like farms in terms of both classification accuracy and computational efficiency. Furthermore, the success of the proposed method signifies that more color features can be discovered in the future research works.

[1]  Pascal Fua,et al.  SLIC Superpixels Compared to State-of-the-Art Superpixel Methods , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Yunhan Dong,et al.  Segmentation and classification of vegetated areas using polarimetric SAR image data , 2001, IEEE Trans. Geosci. Remote. Sens..

[3]  Yi Su,et al.  Region-Based Classification of Polarimetric SAR Images Using Wishart MRF , 2008, IEEE Geoscience and Remote Sensing Letters.

[4]  Abdelhak M. Zoubir,et al.  Segmentation by Classification for Through-the-Wall Radar Imaging Using Polarization Signatures , 2012, IEEE Transactions on Geoscience and Remote Sensing.

[5]  Sven J. Dickinson,et al.  TurboPixels: Fast Superpixels Using Geometric Flows , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Fachao Qin,et al.  Superpixel Segmentation for Polarimetric SAR Imagery Using Local Iterative Clustering , 2015, IEEE Geoscience and Remote Sensing Letters.

[7]  Jean-Claude Souyris,et al.  Support Vector Machine for Multifrequency SAR Polarimetric Data Classification , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[8]  Timo Balz,et al.  Unsupervised polarimetric synthetic aperture radar classification of large-scale landslides caused by Wenchuan earthquake in hue-saturation-intensity color space , 2014 .

[9]  Laurent Ferro-Famil,et al.  Unsupervised terrain classification preserving polarimetric scattering characteristics , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[10]  Ridha Touzi,et al.  Mean-shift and hierarchical clustering for textured polarimetric SAR image segmentation/classification , 2010, 2010 IEEE International Geoscience and Remote Sensing Symposium.

[11]  Moncef Gabbouj,et al.  Semi-Supervised Learning for Ill-Posed Polarimetric SAR Classification , 2014, Remote. Sens..

[12]  Kun-Shan Chen,et al.  A fuzzy neural network to SAR image classification , 1998, IEEE Trans. Geosci. Remote. Sens..

[13]  Jitendra Malik,et al.  Normalized cuts and image segmentation , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[14]  Wenxian Yu,et al.  Superpixel-Based Classification With an Adaptive Number of Classes for Polarimetric SAR Images , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[15]  E. Pottier,et al.  Polarimetric Radar Imaging: From Basics to Applications , 2009 .

[16]  Timo Balz,et al.  Polarimetric Response of Landslides at X-Band Following the Wenchuan Earthquake , 2014, IEEE Geoscience and Remote Sensing Letters.

[17]  Hiroyoshi Yamada,et al.  A four-component decomposition of POLSAR images based on the coherency matrix , 2006, IEEE Geoscience and Remote Sensing Letters.

[18]  Hiroyoshi Yamada,et al.  Four-component scattering model for polarimetric SAR image decomposition , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[19]  Serkan Kiranyaz,et al.  Integrating Color Features in Polarimetric SAR Image Classification , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[20]  Lei Shi,et al.  Polarimetric SAR Image Segmentation Using Statistical Region Merging , 2014, IEEE Geoscience and Remote Sensing Letters.

[21]  Ying-Hua Wang,et al.  PolSAR Image Segmentation by Mean Shift Clustering in the Tensor Space , 2010 .

[22]  S. Süsstrunk,et al.  SLIC Superpixels ? , 2010 .

[23]  Dorin Comaniciu,et al.  Mean Shift: A Robust Approach Toward Feature Space Analysis , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[24]  Rabab Kreidieh Ward,et al.  Segmentation and Classification of Polarimetric SAR Data Using Spectral Graph Partitioning , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[25]  Laurent Ferro-Famil,et al.  Unsupervised classification of multifrequency and fully polarimetric SAR images based on the H/A/Alpha-Wishart classifier , 2001, IEEE Trans. Geosci. Remote. Sens..

[26]  Eric Pottier,et al.  An entropy based classification scheme for land applications of polarimetric SAR , 1997, IEEE Trans. Geosci. Remote. Sens..

[27]  Stefano Soatto,et al.  Quick Shift and Kernel Methods for Mode Seeking , 2008, ECCV.

[28]  Thomas L. Ainsworth,et al.  Unsupervised classification using polarimetric decomposition and the complex Wishart classifier , 1999, IEEE Trans. Geosci. Remote. Sens..