基于图像块分割及差异演化的多聚焦图像融合算法

Multi-focus Image Fusion Based on Image Block Segment and Differential Evolution Algorithm

对于可见光成像系统光学镜头的焦距是有限的,很难将场景中的所有物体都成像清晰.可以对同一场景不同聚焦点的多幅图像进行融合处理,来获取一幅处处清晰的图像.提出了一种基于图像块分割及差异演化的多聚焦图像融合算法,即先把源图像进行分块,再用空间频率作为清晰度评价函数,判断融合子块应取自哪幅源图像,最终重构成新图像.结果表明,与小波变换和遗传算法相比,该方法速度快且融合效果好.

In the visible light imaging system optical lenses are limited in their depth of field. Con- sequently, it is difficult to obtain a good focus for all objects in a picture. One way to get an eve- rywhere-in-focus image is to fuse the images of the same scene taken from different focal settings. In this paper, a novel optimal method for multi-focus image fusion using the image block segment and differential evolution algorithm is presented. The source images are first decomposed into blocks, and then space frequency is employed as a sharpness criterion function to select the sharp- er blocks. The selected blocks are finally combined to construct the fused image. Experimental results prove that this algorithm is faster and more precise than the wavelet transform and genetic algorithm.