结构优化型颜色传递融合方法

Architecture Optimized Version Color Transfer Based Fusion Method

提出一种红外和可见光图像的快速彩色图像融合算法:结构优化型颜色传递融合方法.算法直接用灰度融合图像和源多波段图像的差异信号构成源YCBCR分量,然后在YCBCR空间运用统计颜色传递技术形成一幅具有与目标图像相似色彩效果的彩色融合图像.文中从数学上证明了YUV等符合通用YCBCR空间模型的颜色空间也可用于本文算法,所产生的融合结果与用YCBCR空间得到的结果相同.算法不仅限于融合红外和可见光图像,还可以用于融合其他谱段的双波段图像.实验结果表明,提出的彩色图像融合算法能够有效地生成一幅具有自然日光色彩效果的融合图像,算法中即使采用像素平均法进行灰度融合同样可以获得令人满意的融合效果.

A fast color image fusion algorithm for merging infrared and visible images is presented,which is named Architecture Optimized Version Color Transfer Based Fusion Method.The method directly uses the grayscale fused image and the difference signals of the input multiband images to construct the source YCBCR components,then uses statistical color transfer technique in YCBCR space to form a color fused image with similar color appearance as the target image.The paper mathematically proves that color spaces conforming to the general YCBCR space framework,such as YUV space,can be used as an alternative to YCBCR space in the proposed method.The approach is not only limited to fuse infrared and visible images,but also suitable for merging other dual-band image types.Experimental results show that the proposed color image fusion algorithm can effectively produce a natural appearing daytime-like color fused image,and even using the pixel averaging fusion scheme to implement the grayscale fusion can also provide a satisfactory result.