A fast segmentation approach of SAR image by fusing optical image

Abstract

Optimal segmentation results can be obtained by synthetic aperture radar (SAR) image segmentation based on the Markov Random Field (MRF) model. However, MRF segmentation is very time-consuming because of employing the simulated annealing algorithm to optimize energy function. To speed up SAR image segmentation based on the MRF model, this paper investigates a fast segmentation approach for SAR imagery by fusing optical imagery. First, the optical image is applied to accelerate SAR image segmentation by selecting the uncertain pixels which only attend the SAR image segmentation and marking the certain pixels which don't attend the SAR image segmentation. Second, a fast annealing strategy is proposed to shorten the annealing time under low temperature. Finally, Computer simulation is given to validate the effectiveness of the proposed method.