基于三次B样条提升方案的图像融合算法  

Image Fusion Based on Cubic B-spline Lifting Scheme Transform

作　　者：高丙坤[1] 韩双[1] 栾岩慧[2] 

机构地区：[1]大庆石油学院,黑龙江大庆163318 [2]大庆石化公司通讯中心,黑龙江大庆163318

出　　处：《现代电子技术》2009年第6期113-115,共3页Modern Electronics Technique

摘　　要：提出了一种基于三次B样条提升方案的图像融合算法,该算法是对三次B样条小波基进行提升,采用一定的融合决规则对高频分量和低频分量进行融合。高频分量采用多尺度下邻域方差法来计算,低频分量采取求平均的方法进行融合,最后将三次B样条提升算法与传统的小波变换算法的融合图像和参数指标进行比较,实验表明不论从视觉效果,还是采用信息熵,联合熵和平均梯度作为评价标准,都优于传统的小波变换法。A mathod of image fusion that is the lifting scheme to cubic B- spline wavelet is proposed. The high -frequency coefficients and low - frequency coefficients of the fused image can be obtained by using different fusion rules. The multi - scale neighborhood variance can be used with high frequency coefficients and for the average method can be used with the low -frequency coefficients. Comparing image fusion and parameters separately between lifting algorithms based on cubic B- spline and the wavelet transform tra- ditional. Experiments show that regardless of the vison feelings, or the use of information entropy, connection entropy and average gradient as an evaluation criterion, the method is better than the traditional wavelet transform method.

关 键 词：图像融合 提升小波 三次B样条 融合规则 评价指标 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]