节点文献
人体脸部灰度图像上色的改进算法
Improvement of Colorizing Algorithm for Greyscale Face Image
【摘要】 采用分块多分辨率上色方法 ,首先对低分辨率灰度图像上色 ;然后再对高分辨率图像做颜色变换 基于亮度、邻域信息和空间相关性的相似性测度控制像素的上色 实验结果表明 ,该算法可以改进脸部图像的上色效果 ,且提高了上色速度
【Abstract】 A swatch-based mutli-resolution colorization algorithm is presented. The lower resolution greyscale image is first colorized following the original image, then color transfer is conducted for the higher resolution image. Luminance, neighborhood information and space correlation of pixels are used to control the pixel colorization. Experiments show that the colorizing result of gray face image is improved and the processing time is reduced.
【关键词】 脸部图像;
上色;
多分辨率;
空间相关性;
相似性测度;
【Key words】 face image; colorizing; multi-resolution; space correlation; similarity measure;
【Key words】 face image; colorizing; multi-resolution; space correlation; similarity measure;
【基金】 国家“八六三”高技术研究发展计划 ( 2 0 0 1AA2 3 10 3 1);国家重点基础研究发展规划项目 (G19980 3 0 60 8);国家科技攻关计划课题奥运科技专项 ( 2 0 0 1BA90 4B0 8)资助
- 【文献出处】 计算机辅助设计与图形学学报 ,Journal of Computer Aided Design & Computer Graphics , 编辑部邮箱 ,2004年08期
- 【分类号】TP391.4
- 【被引频次】39
- 【下载频次】282