基于多特征的多聚焦图像融合算法  

Multi-Focus Image Fusion Method Based on Multi-Feature

作　　者：候佳康 贺长秀 

机构地区：[1]四川大学计算机学院 [2]中国人民解放军63615部队

出　　处：《现代计算机（中旬刊）》2016年第9期43-47,共5页Modern Computer

基　　金：国家自然科学基金(No.61071162)

摘　　要：针对基于小波变换的传统图像融合算法在融合规则上考虑特征单一的问题,提出基于多特征的多聚焦图像融合方法。首先对源图像进行N层小波分解,得到高频系数和低频系数。然后对高频系数提出一种综合考虑方差,梯度,能量因素进行融合的方法,对低频系数采用传统的加权平均融合方法。仿真实验结果表明,提出的融合方法与传统的图像融合方法对比,无论是从主观视觉效果还是客观指标(信息熵、空间频率、标准差、平均梯度)都有一定程度提高。The traditional image fusion algorithm is based on wavelet consider the single feature. Therefore, proposes a multi-focus image fusion method based on multi-feature. Firstly, decomposes N layer wavelet of the source image, and obtains the high-frequency coefficients and low-frequency coefficients. Secondly, for high frequency coefficients, proposes a new method for the fusion, which is a comprehensive consideration of the variance, the gradient and the energy. For the low-frequency coefficients, uses the weighted average algorithm. Simu- lation experiment results show that, compared with the traditional image fusion method, the proposed fusion method has a certain degree of improvement in both subjective and objective indicators.

关 键 词：小波变换 图像融合 多特征 算法 图像评价 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]