一种基于Curvelet变换的彩色图像融合算法  

Algorithm for color image fusion based on Curvelet transform

作　　者：高战立 余桂英[1] 

机构地区：[1]中国计量学院计量测试工程学院,浙江杭州310018

出　　处：《中国计量学院学报》2013年第4期419-424,共6页Journal of China Jiliang University

基　　金：浙江省重大科技专项计划项目(No.2011C12025)

摘　　要：针对图像传感器的多聚焦成像问题,提出了一种基于YIQ颜色模型和第二代Curvelet变换的图像融合方法.通过将待融合图像转换到YIQ颜色模型,考虑到图像的颜色各分量相关性和算法的计算复杂度,对亮度分量Y和色度分量I、Q分别采用了不同的融合策略:亮度分量Y进行Curvelet变换,分解后的高频系数采用区域梯度和能量加权相结合的融合准则,低频系数采用能量自适应加权的融合准则,再进行Curvelet逆变换重构得到分量Y,色度分量I、Q分别通过区域信息熵取大的方法得到.将融合得到的三分量进行YIQ逆变换,从而实现多聚焦彩色图像的融合.通过实验对比,算法取得较好的融合效果.Focusing on the multi-focus imaging problem of image sensor devices,a color image fusion method based on the YIQ color model and the second generation curvelet transform was proposed.The original images were transformed in YIQ space.In view of the correlation of each color component and computation complexity,different fusion rules were applied to the Yluminance component and chromaticity component. The Y component was decomposed with Curvelet transform.The coefficients of low-frequency and high- frequency in the transform were obtained.The high-frequency coefficients were gained by region gradient and energy weighted,while the low-frequency coefficients were decided by the energy adaptive weighted fusion rule.The Ycomponent was reconstructed by inverse Curvelet transform.The other components were based on the maximum region information entropy rule.Through the YIQ inverse transform the color muti-focus fusion image was obtained.Experiments showed that the proposed algorithm has good fusion effect.

关 键 词：CURVELET变换 区域梯度和能量加权 区域信息熵 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]