An Unsupervised Classification Method of POLINSAR Image Based on Bayesian Information Criterion

An unsupervised classification algorithm established on the Bayesian Information Criterion (BIC) is presented for Polarimetric and Interferometric SAR (PolInSAR) images. First, an initial classification result is obtained by using Shannon entropy characteristic. Then, the result is optimized by Expectation-Maximization (EM) iteration algorithm and LabelCost optimization algorithm. Meanwhile, the method uses BIC to determine the number of clusters automatically. The experimental results show that the proposed method can not only obtain satisfied classification results, but also automatically determine the number of clusters.