摘要
针对红外与可见光图像融合,提出了一种基于区域能量和修正的视觉特征对比度的低频融合规则,以尽可能多地保留红外图像的热目标信息及可见光图像的光谱信息。实验结果表明,在非下采样轮廓波变换和小波变换域,新的融合规则能都充分利用源图像的互补和冗余信息,使融合图像具有更好的主观视觉效果。
Aiming at infrared and visible image fusion, a low-frequency fusion rule is proposed in this paper based on regional energy and modified visual features, to reserve the hot target information of infrared images and the spectrum information of visible images as much as possible. The experiment results show that, in the nonsubsampled contourlet transform （NSCT） and wavelet transform domain, the novel low-frequency fusion rule can make full use of the complementary and redundant information of the source images, so the fusion images have better subjective visual effects.
出处
《传感器世界》
2013年第5期14-17,共4页
Sensor World
关键词
图像融合
区域能量
视觉特征对比度
非下采样轮廓波变换
小波变换
image fusion
regional energy
visual feature contrast
Nonsubsampled Contourlet Transform （NSCT）
Wavelet Transform（WT）
