SAR image despeckling by employment of multiwavelet based Hidden Markov Model
Wen Long Song, Dong Seog Han · 2016
Despite the high resolution of synthetic aperture radar (SAR) images, speckle noise remains a big challenge in the image. A novel method of restraining speckle noise in a SAR image by use of the multiwavelet based Hidden Markov Model (HMM) is proposed in this paper. In order to estimate the HMM parameters, the expectation maximization (EM) algorithm is adopted. The experimental results show that the proposed despeckling method in this paper improves the image quality and increases the equivalent number of looks (ENL) while keeping more edge information compared to the single wavelet multiscale based HMM.