Speckle Reduction of PolSAR Images in Forest Regions Using Fast ICA Algorithm

Advancement in Polarimetric synthetic aperture radar technology (PolSAR) and its ability to capture images in different polarizations, facilitate accessing large volumes of information from a scene. These images as well as conventional SAR images are degraded by speckle noise. Processing and reducing the speckle of all the images separately is time consuming. So applying methods to simultaneously process different channels information can be very helpful. Several methods have been proposed for speckle reduction and each has advantages and disadvantages. In this paper, a despeckling approach based on fast independent component analysis (Fast ICA) algorithm is proposed for improving of the results when polarimetric channels are added. The results show the more input images to Fast ICA algorithm leads to better separation between signal components and the speckle. Analysis of the results for the ALOS (Advanced Land Observing Satellite) PALSAR polarimetric images showed that combination of polarimetric channels improved 37 % the ENL value in comparison with only using the HH (horizontal-horizontal)-, HV (horizontal-vertical)-, and VV (vertical-vertical)- polarized channels.

[1]  Alexander A. Sawchuk,et al.  Adaptive restoration of images with speckle , 1987, IEEE Trans. Acoust. Speech Signal Process..

[2]  Carlos López-Martínez,et al.  Polarimetric SAR speckle noise model , 2003, IEEE Trans. Geosci. Remote. Sens..

[3]  Haijiang Wang,et al.  Comparision of Two ICA Based Methods for Speckle Reduction of Polarimetric SAR Images , 2009, 2009 Asia-Pacific Conference on Information Processing.

[4]  Paola Ballatore,et al.  Extracting digital elevation models from SAR data through independent component analysis , 2011 .

[5]  Xiaoxia Huang,et al.  A Level Set Filter for Speckle Reduction in SAR Images , 2010, EURASIP J. Adv. Signal Process..

[6]  Lamei Zhang,et al.  A novel super-resolution method of PolSAR images based on target decomposition and polarimetric spatial correlation , 2011 .

[7]  Zequn Guan,et al.  A novel remote sensing image fusion method based on independent component analysis , 2011 .

[8]  Yiming Pi,et al.  The fast fixed-point algorithm for speckle reduction of polarimetric SAR image , 2005 .

[9]  James V. Stone.  Applications of ICA , 2004 .

[10]  Alexander A. Sawchuk,et al.  Adaptive Noise Smoothing Filter for Images with Signal-Dependent Noise , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Alin Achim,et al.  SAR image denoising via Bayesian wavelet shrinkage based on heavy-tailed modeling , 2003, IEEE Trans. Geosci. Remote. Sens..

[12]  E. Oja,et al.  Independent Component Analysis , 2013 .

[13]  Saïd Kharbouche,et al.  Speckle reducing in PolSAR images for topographic feature extraction , 2013 .

[14]  Josaphat Tetuko Sri Sumantyo,et al.  Employing a Method on SAR and Optical Images for Forest Biomass Estimation , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[15]  Gui Gao,et al.  Statistical Modeling of SAR Images: A Survey , 2010, Sensors.

[16]  Huanxin Zou,et al.  Feature selection and weighted SVM classifier-based ship detection in PolSAR imagery , 2013 .

[17]  E. Nezry,et al.  Adaptive speckle filters and scene heterogeneity , 1990 .

[18]  A. Antico,et al.  Independent component analysis of MODIS-NDVI data in a large South American wetland , 2012 .

[19]  Afzal Godil,et al.  Investigating the Bag-of-Words Method for 3D Shape Retrieval , 2010, EURASIP J. Adv. Signal Process..

[20]  Jianya Gong,et al.  A decomposition-free scattering mechanism classification method for PolSAR images with Neumann’s model , 2013 .

[21]  Victor S. Frost,et al.  A Model for Radar Images and Its Application to Adaptive Digital Filtering of Multiplicative Noise , 1982, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[22]  H. Arsenault,et al.  Properties of speckle integrated with a finite aperture and logarithmically transformed , 1976 .

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

[24]  J. Yackel,et al.  Comparing matrix distance measures for unsupervised POLSAR data classification of sea ice based on agglomerative clustering , 2013 .

[25]  Jong-Sen Lee,et al.  Digital Image Enhancement and Noise Filtering by Use of Local Statistics , 1980, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[26]  Jong-Sen Lee,et al.  Principal components transformation of multifrequency polarimetric SAR imagery , 1992, IEEE Trans. Geosci. Remote. Sens..

[27]  FuYusheng,et al.  THE FAST FIXED-POINT ALGORITHM FOR SPECKLE REDUCTION OF POLARIMETRIC SAR IMAGE , 2005 .

[28]  Scott T. Acton,et al.  Speckle reducing anisotropic diffusion , 2002, IEEE Trans. Image Process..

[29]  Shi-qi Huang,et al.  A novel method for speckle noise reduction and ship target detection in SAR images , 2009, Pattern Recognit..

[30]  Yiming Pi,et al.  Applications of ICA for the enhancement and classification of polarimetric SAR images , 2008 .

[31]  Jong-Sen Lee,et al.  Speckle reduction in multipolarization, multifrequency SAR imagery , 1991, IEEE Trans. Geosci. Remote. Sens..

[32]  LeeJong-Sen.  Digital Image Enhancement and Noise Filtering by Use of Local Statistics , 1980 .

[33]  Carlos López-Martínez,et al.  Filtering and Segmentation of Polarimetric SAR Data Based on Binary Partition Trees , 2012, IEEE Transactions on Geoscience and Remote Sensing.

[34]  S. Chitroub,et al.  Independent Component Analysis of POLSAR Images. Relative Newton-Based Approach , 2007, 2007 IEEE International Conference on Signal Processing and Communications.

[35]  Seungjin Choi,et al.  Independent Component Analysis , 2009, Handbook of Natural Computing.

[36]  Torbjørn Eltoft,et al.  A Stationary Wavelet-Domain Wiener Filter for Correlated Speckle , 2008, IEEE Transactions on Geoscience and Remote Sensing.