基于内容的图像染色方法  

Image Color Transfer Method Based on Content

作　　者：汲业[1,2] 陈燕[1] 牟向伟[1] 屈莉莉[1] 

机构地区：[1]大连海事大学交通运输管理学院,辽宁大连116023 [2]大连灵动科技发展有限公司,辽宁大连116023

出　　处：《计算机工程》2012年第20期197-199,共3页Computer Engineering

基　　金：国家自然科学基金资助项目(70940008)

摘　　要：利用彩色图像对灰度图像进行染色时,色源图像质量的优劣是染色成功的关键。为此,提出基于纹理和颜色特征的图像染色方法。提取图像视觉内容特征中的纹理和色彩作为索引,计算色源图像与灰度图像的纹理相似度。搜索内容与灰度图像相近的色源图像,以提高染色的成功率。提取色源图像的颜色特征进行像素采样,以增强染色效果。实验结果表明,该方法的染色成功率较高,染色效果较好。When transferring color from colorful image to grayscale image,a suitable source image is a key to succeed rendering.Thus,a color transfer method based on texture and color feature extraction is introduced in this paper.Features of texture and color are extracted as indexes,the texture similarity between source images and grayscale image is calculated to retrieve a similar source image to improve the color transfer,and the color feature is extracted to be used to select sample pixels from the source image to improve the rendering effect.Experimental results show that the method can improve the probability of color transfer with better performance.

关 键 词：染色 纹理特征 颜色特征 灰度图像 色源图像 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]