多聚焦图像融合方法仿真研究  

Multi-focus Image Fusion Simulation

作　　者：鲍建成[1] 丁伟[2] 

机构地区：[1]江苏海事职业技术学院信息工程系,江苏南京211170 [2]南阳理工学院,河南南阳473004

出　　处：《计算机仿真》2011年第12期240-243,共4页Computer Simulation

摘　　要：研究了多聚焦图像融合问题。针对焦点不同的图像融合后,在传输过程中由于受到不同的噪声的干扰造成结果清晰度不高等问题,为了提高图像的清晰度,提出了一种改进的多小波变换多聚焦图像融合算法。首先利用不同的小波对待融合图像进行小波分解,采用多尺度双结构元对输入图像进行滤波,在融合规则选择中,采用加权平均法选择高频系数,采用绝对值取大方法选择低频系数。对多组多聚焦图像进行实验,实验表明采用该方法能够更好的保留图像边缘信息,融合效果明显优于传统的图像融合方法。In the transmitting process,low definition may be caused by noises.In order to improve the image clarity,an improved multi-focus image fusion algorithm based on multi-wavelet transform was proposed.By using different wavelets,the decomposition were carried out with the images to be fused,and multi-scale and double structuring element was used in filtering of the input image.For the fusion rules,the weighted average method was used to choose the high frequency coefficients,and the largest absolute value was used for the low-frequency coefficients.The experiments of multiple sets of multi-focus images show that this method can be used to retain edge information with better effects,and the image fusion results are better than the traditional image fusion methods.

关 键 词：图像融合 边缘检测算子 多小波变换 熵值 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]