Unsupervised polarimetric SAR image classification using fisher linear discriminant

How to use the POLSAR data to classify and interpret the conditions of the earth is a very important research field of POLSAR. In this paper, we propose an improved algorithm on the basis of studying and analyzing some common algorithms. This technique introduces the fisher criterion in the feature selection and the hierarchical method in the classification of POLSAR image, which can improve Lee´s combination method of the unsupervised classification based on polarimetric target decomposition and the maximum likelihood classifier based on the complex Wishart distribution. The effectiveness of this algorithm is demonstrated using a polerimetric SAR image.