基于多层次MRF的多分辨率图像融合算法

Multiresolution Image Fusion Based on Hierarchical Markov Random Field Model

图像数据融合的目的是恢复出分布在不同图像中的有用信息。在多层次MRF模型的基础上提出了一种多分辨率图像融合算法。该算法将定义在多层次图结构上的非线性因果Markov模型与贝叶斯SMAP(sequenti almaximumaposteriori)准则结合起来 ,弥补了MAP(maximumaposteriori)准则在多层次图结构上计算不合理的缺陷。实验部分中 ,对两种算法用于被高斯白噪声污染的合成图像的恢复结果进行了比较 ,并将该算法用于实际的多分辨率航空图像。实验结果表明了该算法的优越性。

Image data fusion is the technique that can recover the hidden information exist in different images. A multiresolution image fusion algorithm based on hierarchical Markov random field model is proposed in this paper. This algorithm combines nonlinear causal Markov model defined on hierarchical graph structures with Bayesian SMAP criterion, and circumvents the drawback of classical MAP criterion on the hierarchical graph structures. The experiments on synthetic images and real aerial images using this algorithm and MAP criterion indicate the advantage of this algorithm.\;