Satellite Image Restoration Based on HMT-ForWaRD Method
Xingming Long · 2005
The satellite image restoration is an ill-condition problem. In this paper the degraded satellite images are restored by our suggested HMT ForWaRD method, in which the degraded image is deblurred by Wiener filter in Fourier domain and then the remained unwanted white gaussian noises are removed by hidden Markov tree (HMT) method in the complex wavelet domain. The simulation is done for solving the typical convolution and noised linear degraded satellite image and its results show that the suggested satellite image restoration method performs better than common Wiener restoration method by visual feelings and objective criteria.