摘要
利用Contourlet变换的多尺度、局部化、方向性和各向异性等优点,提出了一种基于区域特性的Contourlet域多聚焦图像融合算法.该算法将源图像分解至Contourlet变换域,在不同尺度、不同方向的子带中结合区域特性进行图像融合,低频和高频子带中分别采用区域方差和区域能量作为融合规则,最后通过反变换得到融合图像.实验结果表明,所提出的算法能够更好地提取原始图像特征,融合后的图像具有更好的主观视觉效果,与经典的梯度金字塔算法和小波变换算法相比,新算法的均方误差最大值仅为前二者的34.8%和42.6%.
Utilizing the contourlet＇s advantages of multiscale, localization, directionality and anisotropy, a multifocus image fusion algorithm based on region statistics in contourlet domain is developed. Source images are firstly decomposed to the domain of the contourlet transform, the image fusion is then implemented in subbands with different scale and direction combining with region statistics. Regional variance and local energy are adopted as fusion rules in lowpass and high- pass subbands, respectively. Finally the fused image is obtained through inverse transform. Experimental results show that the proposed method can detect image features more effectively and the fused image has better subjective visual effect. Compared to traditional gradient pyramid algorithm and wavelet based algorithm, the maximum of mean square error of the proposed algorithm is only 34. 8% and 42.6% of the formers, respectively.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2007年第4期448-452,共5页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(60572152)
陕西省自然科学基金资助项目(2005F26)
