计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (25): 63-65.

• 学术探讨 • 上一篇    下一篇

基于聚类的自动颜色传输

王常亮1,2   

  1. 1.东南大学 计算机科学与工程学院,南京 210096
    2.浙江工业职业技术学院 计算机系,浙江 绍兴 312000
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-09-01 发布日期:2007-09-01
  • 通讯作者: 王常亮

Automatic color transfer based on clustering

WANG Chang-liang1,2   

  1. 1.School of Computer Science and Engineering,Southeast University,Nanjing 210096,China
    2.Zhejiang Industry Vocational and Technical College,Shaoxing,Zhejiang 312000,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-09-01 Published:2007-09-01
  • Contact: WANG Chang-liang

摘要: 提出了一种基于聚类的颜色传输算法。利用算法可以自动完成彩色图像间、彩色图像和黑白图像间的全局或者局部颜色传输。首先,对于源图像,将采用混合高斯模型进行建模,并利用EM算法求得相应参数完成图像自动聚类。对于目标图像,如果是彩色图像,采用同样聚类方法。如果是黑白图像,采用K-Means聚类法。完成源图像和目标图像的聚类,并建立双方聚类区域的对应关系后,通过改变目标图像中像素的Lαβ通道值完成色彩传输功能。实验证明,算法高效且无需人工干预,颜色传输效果很理想。

关键词: 颜色传输, 聚类, 高斯混合模型

Abstract: A new clustering-based algorithm for color transfer is proposed in this paper.Global or local color transfer between color images or between color image and gray-scale image can be performed by this algorithm.Firstly,GMM-EM algorithm is jointly applied to automatically group color source image into regions.The same GMM-EM algorithm is used if target image is color image and k-means algorithm is used in case of gray-scale image.Then,corresponding map will be built between source image regions and target image regions after clustering.Finally,Lαβ values of pixel in the target image are changed to impose domain color of source image on target image.The experiments show the presented algorithm is highly automatically and effective and,satisfying results can be obtained.

Key words: color transfer, clustering, Gaussian mixture model