基于灰值形态学的不同聚集图像融合方法研究  

Study on Fusion Multifocus Images Using Gray-level Morphology

作　　者：孙玉秋[1] 李亮[1] 

机构地区：[1]长江大学信息与数学学院,湖北荆州434023

出　　处：《长江大学学报（自科版）（上旬）》2010年第4期83-86,共4页JOURNAL OF YANGTZE UNIVERSITY (NATURAL SCIENCE EDITION) SCI ＆ ENG

基　　金：国防预研基金项目(51483020104JW05)

摘　　要：提出了一种基于灰值形态学变换的多聚焦图像融合方法。针对不同聚焦的灰度图像,对其进行灰值形态学变换,先进行形态开运算再进行形态闭运算,然后对其采样,重复该过程可以得到不同大小的子图像;在相同的图像分解级上采用局部灰度值最大准则融合相应的子图像,对融合子图像先进行形态闭运算再进行开运算,然后降采样,最后重构得到融合结果图像。并将融合结果图像与标准测试图像进行了分析对比,对比结果表明,对不同聚焦灰度图像进行灰值形态学运算,所得到的图像融合效果好,运算效率高。A method is developed for fusion of two spatially registered images with differing focus points.Gray-level morphology transforms are operated on these differing focus images according to their features.Firstly morphological open operation is performed and morphological close operation is followed,then sample is collected.After repeating this procedure,some sub-images with different sizes can be obtained.To fuse these sub-images at the same decomposition level by local maximum fusion rule and reconstruct them,fused images can be obtained.Comparing fused resulting images with standard corrected images,the experimental results show that the method using morphology to fuse multifocus images is better and its efficiency is high.

关 键 词：灰度图像 数学形态学 灰值形态学 融合 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]