Usually, Wishart H/α/A classification is an effective unsupervised classification method. However, the anisotropy
parameter (A) is an unstable factor in the low signal noise ration (SNR) areas; at the same time, many clusters are useless
to manually recognize. In order to avoid too many clusters to affect the manual recognition and the convergence of
iteration and aiming at the drawback of the Wishart classification, in this paper, an enhancive unsupervised Wishart
classification scheme for POLSAR data sets is introduced. The anisotropy parameter A is used to subdivide the target
after H/α classification, this parameter has the ability to subdivide the homogeneity area in high SNR condition which
can not be classified by using H/α. It is very useful to enhance the adaptability in difficult areas. Yet, the target
polarimetric decomposition is affected by SNR before the classification; thus, the local homogeneity area's SNR
evaluation is necessary. After using the direction of the edge detection template to examine the direction of POL-SAR
images, the results can be processed to estimate SNR. The SNR could turn to a powerful tool to guide H/α/A
classification. This scheme is able to correct the mistake judging of using A parameter such as eliminating much
insignificant spot on the road and urban aggregation, even having a good performance in the complex forest. To
convenience the manual recognition, an agglomerative clustering algorithm basing on the method of deviation-class is
used to consolidate some clusters which are similar in 3by3 polarimetric coherency matrix. This classification scheme is
applied to full polarimetric L band SAR image of Foulum area, Denmark.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
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.