Water Change Detection Based on Complex Wishart Distribution Model Using Polarimetric SAR Image

Inhalt:
Since the covariance matrix of single polarimetric SAR image obeys complex Wishart distribution, the covariance matrix which describes the joint image of two different phases of polarimetric SAR images also obeys complex Wishart distribution. The joint image constructed by complex Wishart distribution contains three characteristics: First, noise is significantly suppressed. Second, the unchanged water shows significant dark color compared with others because of water scattering characteristics. Third, the ratio image acquired by the single image and the joint image combined with a threshold can divide the water level change into four categories: land into water bodies, semi- flooded area into water bodies, water bodies into land and semi- flooded area into land. It shows more advantages compared with the complex Wishart likelihood-ratio test method which only divides the water level change into two categories. The experimental results show that this method can obtain the accurate detection results.