基于人工鱼群优化分块的多聚焦图像融合

在多聚焦图像的融合过程中,对源图像采用固定大小的分块会导致融合后的图像存在块效应、边缘模糊甚至聚焦错误。为了克服此问题,提出了一种新的基于人工鱼群优化分块的多聚焦图像融合方法。首先,将源图像分解成互不重叠的方块,利用聚焦准则选取清晰度高的方块,将已选择的方块合并重构成初始融合图像。然后,利用改进的人工鱼群优化算法,根据一定的适应度值,寻找最优大小的分块方式,获得更优的融合图像。该方法与基于空域、频域及其他优化算法的融合方法进行了多个实验比较,结果表明,该方法获得的融合图像具有较好的客观质量和主观视觉感觉。

Multi-focus images fusion based on block optimization using artificial fish-swarm algorithm

The fixed block size of source images will result in blocking artifacts, fuzzy edge and focus error in multi-focus image fusion. To solve this problem, a new multi-focus image fusion algorithm based on block optimization using artificial fish-swarm is proposed. Firstly, the source images are decomposed into non-overlapping blocks and the sharper blocks are selected using a sharpness criterion. The selected blocks are combined to construct the initial fused image. Then, an improved artificial fish-swarm algorithm is used to optimize the block size according to a fitness function. The final fused image is obtained based on the best block size. Experimental results show that the proposed fusion method has a good quantitative evaluation and visual effect compared to other traditional methods.