计算机工程与应用 ›› 2021, Vol. 57 ›› Issue (21): 216-223.DOI: 10.3778/j.issn.1002-8331.2010-0368

• 图形图像处理 • 上一篇    下一篇

光照不均匀图像的自适应增强算法

汤子麟,刘翔,张星   

  1. 1.上海工程技术大学 电子电气工程学院,上海 201620
    2.上海工程技术大学 管理学院,上海 201620
  • 出版日期:2021-11-01 发布日期:2021-11-04

Adaptive Enhancement Algorithm for Non-uniform Illumination Images

TANG Zilin, LIU Xiang, ZHANG Xing   

  1. 1.School of Electrical and Electronic Engineering, Shanghai University of Engineering Science, Shanghai 201620, China
    2.School of Management Studies, Shanghai University of Engineering Science, Shanghai 201620, China
  • Online:2021-11-01 Published:2021-11-04

摘要:

针对光照不均匀场景,提出了一种自适应图像增强算法。根据Retinex理论,采取中心环绕法,利用高斯连续卷积来提取场景的光照分布情况。同时,统计输入图像低亮度区域的大小。构造了一种自适应伽马矫正函数,取光照分布情况与低亮度区域内亮度值中位数的比值作为参数,对图像进行伽马校正。高光照区域参数大于1,对亮度起抑制作用,低光照区域参数小于1,对亮度起增强作用。将顶帽变换后图像和伽马矫正后的图像叠加。顶帽变换可以提升图像的全局对比度,伽马函数可以保留细节信息。两者结合后,可以兼顾图像的全局特性和局部细节信息。视觉感受和客观实验指标表明,与参照算法相比,该算法针对不均匀光照图像增强效果显著。

关键词: 图像增强, 不均匀光照, Retinex理论, 伽马矫正函数, 顶帽变换

Abstract:

An adaptive image enhancement algorithm for non-illumination is proposed. According to the center surround method, using Gaussian kernel convoluted repeatedly to extract light of scene. Meanwhile, the size of the low-brightness area in scene is calculated. An adaptive Gamma correction function is constructed to take the ratio of light to the median brightness value in the low-brightness area as a parameter to correct the image. It is greater than 1 in high-light area, algorithm inhibits the brightness, and it is less than 1 in low-light area parameter, algorithm enhances the brightness. Top-Hat transformation is added to the result of Gamma correction. The Top-Hat transformation enhances the overall contrast of the image and the Gamma function preserves detail information. Visual perception and objective experimental indicators show that compared with the known algorithm, the proposed algorithm has a significant effect on non-uniform illumination images.

Key words: image enhancement, non-uniform illumination, Retinex theory, Gamma correction function, Top-Hat transformation