计算机工程

• 图形图像处理 • 上一篇    下一篇

基于金字塔分解的自动色彩均衡算法

张 明,白松浩,邢姗姗   

  1. (中国人民解放军95899 部队,北京100085)
  • 收稿日期:2013-11-29 出版日期:2015-02-15 发布日期:2015-02-13
  • 作者简介:张 明(1981 - ),男,博士后,主研方向:图像处理,视频压缩;白松浩,教授、博士;邢珊珊,工程师、硕士。

Automatic Color Equalization Algorithm Based on Pyramid Decomposition

ZHANG Ming,BAI Songhao,XING Shanshan   

  1. (95899 Unit of PLA,Beijing 100085,China)
  • Received:2013-11-29 Online:2015-02-15 Published:2015-02-13

摘要: 针对传统自动色彩均衡算法运算速度慢、暗区细节不明显的缺点,提出一种基于金字塔分解的自动色彩均衡算法。使用对数运算将暗区图像映射到更适合人眼观察的颜色空间,利用高斯卷积核构造图像金字塔图像序列,从金字塔最顶层图像开始进行自动色彩均衡,并对增强结果逐层进行细化,直至金字塔最底层得到最终的增强图像,细化时只需要少量像素间的比较操作,因而大幅降低了运算复杂度。实验结果表明,该算法能有效改善图像质量,保持图像细节信息,并且计算复杂度较低,便于实际应用。

关键词: 图像增强, 自动色彩均衡, 金字塔分解, retinex 理论, 色彩恒常性, 对数运算

Abstract: An Automatic Color Equalization(ACE) algorithm based on pyramid decomposition is proposed to solve the problems emerge after using traditional ACE,such as low speed,obscure details in dark region. The algorithm utilizes the logarithmic calculation to adjust the image obscure region and employs Gaussian kernel to build image pyramid. Color equalization algorithm is applied at the top level of the image pyramid and the result is iteratively refined layer by layer with comparison between few pixels. The simulation results show that the proposed algorithm can efficiently improve visual quality of the image,preserve the details,reduce the computational complexity and make it more suitable for practical applications.

Key words: image enhancement, Automatic Color Equalization(ACE), pyramid decomposition, retinex theory, color constancy, logarithmic operation

中图分类号: