计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (9): 144-150.DOI: 10.3778/j.issn.1002-8331.1801-0213

• 图形图像处理 • 上一篇    下一篇

自适应的低照度图像增强变分模型

赵斌红,马  帅   

  1. 西安电子科技大学 数学与统计学院,西安 710126
  • 出版日期:2019-05-01 发布日期:2019-04-28

Adaptive Variational Model for Low Light Image Enhancement

ZHAO Binhong, MA Shuai   

  1. School of Mathematics and Statistics, Xidian University, Xi’an 710126, China
  • Online:2019-05-01 Published:2019-04-28

摘要: 针对低照度图像具有低对比度、强噪声等问题,提出了一种自适应的低照度图像增强变分模型。根据亮度分量初步估计低照度图像取反之后图像的透射率,并利用Retinex算法进行细化,以丰富图像的细节。为了抑制噪声的放大且保持边缘信息,根据亮通道先验原理和局部方差构建权重,自适应地调节正则化参数。采用交替迭代最优化方法求解包含透射率和恢复图像的能量泛函得到最优解。实验结果表明,该模型可有效地增强低照度图像,且能保留更多的图像细节、抑制噪声放大,相比于[l1]范数正则化方法,图像尺寸越大,该模型计算效率越高,计算时间优势越明显。

关键词: 低照度图像, 透射率, 亮通道先验, 自适应, 交替迭代最优化

Abstract: The low light images have very low contrast and strong noise. To solve the problem, an adaptive variational model for low light image enhancement is proposed. The transmission of the low light image after inversion is rough estimated based on the luminance component, and the transmission is further detailed by using the Retinex algorithm in order to retain more image details. In order to suppress the amplification of the noise and keep the edges, the regularization parameters are adaptively adjusted according to the priori of the bright channel and the local variance. The optimal solution of energy functional equation including transmittance and restored image is obtained by alternating iterative optimization method. Experimental results show that the proposed model can effectively enhance low light image, retain more details, and suppress noise amplification. Compared with[l1]norm regularization method, the larger the image size, this model has higher computational efficiency and more obvious advantage of computing time.

Key words: low light image, transmission, bright channel prior, adaptive, alternating iterative optimization