基于小波和色彩传递的夜视图像彩色融合技术

摘要

This paper describes a color fusion of night vision image based on wavelet transformation and color transfer that helps observers to get more abundant scene information. Concerning their characteristics, low-level light visible and IR images are combined into a pseudo-color image (source image) using NRL. Then the wavelet transformation is used to decompose the pseudo-color image and a colorful day-time reference image into multi-scale sub-band images. The mean and standard deviation of each sub-band image are calculated and then its components are scaled by the variation ratio of a day-time color target image to the source image. The color distribution of the target image is transferred into the source image. Comparing with traditional linear color transfer, the proposed method can make a color night vision image more realistic and natural, improving the situation perception.%提出一种基于小波和色彩传递的夜视图像的彩色融合方法,使观察者更容易获取图像的场景信息.结合红外图像和微光图像各自的特点,利用NRL法把红外图像和微光图像映射到彩色空间形成假彩色图像(源图像),然后通过小波变换对源图像和参考图像进行多分辨率分解,计算不同分辨率下的均值和标准方差,将各分量根据参考图像和源图像的标准方差比进行缩放,把参考图像的色彩分布传递给源图像.实验结果表明,通过与传统方法的比较,本文算法不仅获得真实场景色彩,还可以提高图像的细节信息,改善场景感知.