局部稀疏编码的自然灰度图像着色方法

Natural Grayscale Image Colorization via Local Sparse Coding

基于自然图像中灰度纹理与颜色信息的正相关性,提出一种基于局部稀疏编码的自然灰度图像着色方法.首先依据参考彩色图像训练得到"亮度-纹理-颜色"联合字典;然后利用纹理特征子空间和相邻像素的双重局部性求解目标灰度图像块的颜色系数,对图像块进行着色;最后通过金字塔方法逐层优化避免颜色突变.实验结果表明,文中方法可以获得颜色真实度高且整体过渡平滑的目标彩色图像,着色过程也更加自动化.

Colorization is a process of adding color to grayscale image under certain constraints. In thi paper we present a new colorization approach via local sparse coding. Our method is based on th positive correlation between grayscale texture and color. We firstly train a joint dictionary from th S e e reference color image, and then solve the coefficients under the locality constraints both in texture subspace and neighboring pixels. To preserve color consistency in local region, we refine the colorized result by image pyramid. Experimental results demonstrate that the proposed method outperforms the previous by providing better reality and smoothness in color field.