摘要
针对灰度可见光与红外图像融合,提出一种基于改进型非下采样轮廓波变换(NSCT)的图像融合方法.不同于经典NSCT模型,改进型NSCT变换摈弃了细节捕捉能力不强的非下采样金字塔分解机制,采用冗余提升不可分离小波变换实现对源图像的多尺度分解;然后,分别采用基于区域平均能量匹配度、邻域系数差和信息熵的融合规则,得到融合图像的低频系数和高频系数;最后,通过改进型NSCT逆变换得到了融合图像.实验结果验证了该方法的有效性.
To the fusion of gray-scale visible light and infrared images,an image fusion method based on improved non-subsampled contourlet transform（NSCT） is proposed.Unlike the classic NSCT model,the improved NSCT substitutes redundant lifting non-separable wavelet（RLNSW） for the former NSP decomposition mechanism whose performance of capturing the detail information is not good to realize the multi-scale decomposition of the source images.Then,the low-frequency coefficients and high-frequency coefficients for the fused image can be obtained by means of the fusion rule based on region mean energy matching degree,neighborhood coefficient deviation and information,entropy,respectively.Finally, the fused image is reconstructed by the inverse improved NSCT.Experimental results show the effectiveness of the proposed method.
出处
《控制与决策》
EI
CSCD
北大核心
2010年第11期1607-1612,共6页
Control and Decision
基金
国家自然科学基金项目(60773209)
陕西省自然科学基金项目(2006F18)
关键词
图像融合
冗余提升不可分离小波
非下采样轮廓波变换
平移不变性
Image fusion
Redundant lifting non-separable wavelet
Non-subsampled contourlet transform
Shift-invariance
