融合参数对对比度塔形分解图像融合方法性能的影响研究

Study on the influence of fusion parameters on the performance of contrast pyramid decomposition-based image fusion scheme

给出了一种基于对比度塔形分解的分层图像融合新方法,其在融合过程中可有选择地突出被融合图像的对比度信息与细节信息,以求达到良好的融合效果。此外,利用熵、交叉熵、互信息量、均方根误差、峰值信噪比等参量,定量评价和研究了不同融合层数、融合规则以及特性区域大小等参数对该融合方法性能的影响。多组不同类型图像的融合实验表明,对融合性能的定量评价结果与融合后图像的视觉效果相吻合。

A novel hierarchical image fusion method based on contrast pyramid decomposition is presented. In order to obtain a better fusion result, the contrast details of the images to be fused can be signalized selectively in the fusion processing. In addition, with the use of the parameters such as entropy, cross entropy, mutual information, root mean square error and peak-to-peak signal-to-noise ratio, the influence of fusion parameters such as different fusion levels, fusion rules and the sizes of salience area on the performance of the fusion scheme is evaluated quantitatively and studied. The experimental results of sets of different images show that quantitative evaluation result for the fusion performance is in accord with the visual perception effect of the fused images.