摘要
针对多聚焦图像,提出了一种基于图像块分割的图像融合方法.采用块区域局部对比度将多聚焦图像分割成三部分:聚焦清晰区域、聚焦模糊区域以及两者之间的边界区域.对于清晰区域和模糊区域,直接选取清晰块区域作为融合后相应块区域;对于边界区域,建立了基于对比度的像素选取方法进行融合处理.实验对比结果表明,该方法明显优于拉普拉斯金字塔融合算法和离散小波变换融合算法.
According to the contrast of block, the original images are separated into clear part, blurring part and the part between clear part and blurring one. For the clear part and blurring one, the former is selected as the corresponding block of fusion image. The part between clear part and blurring one is fused by the choice of the corresponding pixel based on the contrast. The experiments show that the proposed method outperforms the methods based on the Laplacian pyramid and discrete wavelet transform.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2003年第11期1743-1746,1750,共5页
Journal of Shanghai Jiaotong University
基金
国家高技术研究发展计划(863)项目(2001AAl35091)
航空科学基金(02D57003)
上海市科技攻关重大项目(02DZ15001)
上海市科技发展基金项目(015115038)
