首页 | 本学科首页   官方微博 | 高级检索  
     

基于方向性对比度金字塔的图像融合方法
引用本文:刘斌,董迪,陈俊霖.基于方向性对比度金字塔的图像融合方法[J].量子电子学报,2017(4):405-413.
作者姓名:刘斌  董迪  陈俊霖
作者单位:1. 湖北大学计算机与信息工程学院,湖北 武汉,430062;2. 武汉工程大学电气信息学院,湖北 武汉,430205
基金项目:National Natural Science Foundation of China(国家自然科学基金;61471160),Key Project of Natural Science Foundation of Hubei Province(湖北省自然科学基金重点项目
摘    要:基于对比度塔形分解(CP)的图像融合方法具有良好的物理意义,却没有强调方向性的不足,为此提出了一种具有方向性的对比度金字塔图像融合方法.对多聚焦图像进行对比度塔形分解,利用方向滤波器组对高频加方向,得到不同方向的高频子分量.根据不同频率域特点,采用低频分量系数取加权平均、高频分量系数绝对值取大的融合规则,对分解后的子图像进行融合.结果表明:用提出方法得到的融合图像有较高的清晰度和空间分辨率.与基于CP和基于离散小波变换(DWT)的融合方法相比,提出的方法既能保持对比度的含义,又可提供2n个方向信息.

关 键 词:图像处理  图像融合  对比度塔形分解  方向滤波器组

Image fusion method based on directional contrast pyramid
LIU Bin,DONG Di,CHEN Junlin.Image fusion method based on directional contrast pyramid[J].Chinese Journal of Quantum Electronics,2017(4):405-413.
Authors:LIU Bin  DONG Di  CHEN Junlin
Abstract:The image fusion method based on contrast pyramid decomposition has good physical meaning,but there is no emphasis on direction.An image fusion method based on directional contrast pyramid decomposition is proposed.The contrast pyramid decomposition of multi-focus images is performed.Multiple directional sub-images are obtained for each high frequency image by using directional filter banks.According to the characteristics of different frequency domain,the decomposed sub-images are fused,and the fusion rule is that the weighted average value is selected for low frequency sub-band coefficients,while for high frequency sub-band coefficients,the coefficients with larger absolute value are employed.Results show that the fused images obtained by the proposed method have high definition and spatial resolution.Compared to the fusion methods based on the contrast pyramid (CP) decomposition and discrete wavelet transform (DWT),the proposed method can not only keep the meaning of contrast,but also provide 2n direction information.
Keywords:image processing  image fusion  contrast pyramid decomposition  directional filter banks
本文献已被 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号