An Infrared and Visible Image Fusion Algorithm Based on MAP

This paper deals with infrared and visible image fusion problems by maximum a posteriori probability (MAP) estimation. We use imaging mode to construct the conditional probability distribution, assume the fusion image approximate the visible image, and treat sparse property of image gradient as fusion image prior probability distribution. According Bayesian theorem, the fusion image's posterior probability distribution is deduced. The fusion results are obtained by maxim the posterior probability distribution. In experiments, we conduct subjective and objective evaluation. The comparisons show that the MAP-based image fusion method has better performance in both subjective and objective evaluations. The MAP-based image fusion method can be applied to image interpretation, detection and recognition tasks.