Polarimetric SAR image classification based on kernel sparse representation

Xiao Wang; Lamei Zhang; Bin Zou; Zhijun Qiao

doi:10.1117/12.2309523

14 May 2018 Polarimetric SAR image classification based on kernel sparse representation

Xiao Wang, Lamei Zhang, Bin Zou, Zhijun Qiao

Proceedings Volume 10658, Compressive Sensing VII: From Diverse Modalities to Big Data Analytics; 106580L (2018) https://doi.org/10.1117/12.2309523
Event: SPIE Commercial + Scientific Sensing and Imaging, 2018, Orlando, FL, United States

Abstract

Polarimetric SAR obtains rich target scattering information by utilizing different polarizations to transmit and receive radar signals alternately, which has become an important tool for ground exploration. Presently, there are still some problems about the classification of PolSAR image because of the nonlinear data. Nonlinear features often lead the data difficult to distinguish in the conventional dimensions. Kernel method maps data to high-dimensional space, making the linearly inseparable data in the original dimension can be linearly separated in the high-dimensional space. Based on the study of the features of PolSAR data and signal sparse representation theory, this paper proposes a PolSAR image classification method based on kernel sparse representation, which optimizes the polarimetric and spatial information in PolSAR data, uses the kernel function method to solve the adverse effect of the nonlinear features on the classification results in the PolSAR image to obtain more accurate classification results. The experiment uses the fully polarimetric SAR data in San Francisco in the United States obtained by airborne AIRSAR, the advantages of kernel sparse representation in PolSAR image classification can be seen from the results.

Conference Presentation

Citation Download Citation

Xiao Wang, Lamei Zhang, Bin Zou, and Zhijun Qiao "Polarimetric SAR image classification based on kernel sparse representation", Proc. SPIE 10658, Compressive Sensing VII: From Diverse Modalities to Big Data Analytics, 106580L (14 May 2018); https://doi.org/10.1117/12.2309523

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available