光子学报 ›› 2019, Vol. 48 ›› Issue (5): 510001-0510001.doi: 10.3788/gzxb20194805.0510001

• 图像处理 • 上一篇    下一篇

基于双域分解的矿井下图像增强算法

田子建1, 王满利1,2, 吴君2, 桂伟峰2, 王文清3   

  1. 1. 中国矿业大学(北京) 机电与信息工程学院, 北京 100083;
    2. 河南理工大学 物理与电子信息学院, 河南 焦作 454000;
    3. 北京工业职业技术学院, 北京 100042
  • 收稿日期:2018-12-04 出版日期:2019-05-25 发布日期:2019-03-18
  • 通讯作者: 田子建(1964-),男,教授,博导,主要研究方向为信息与通信技术.Email:Tianzj0726@126.com
  • 基金资助:

    国家自然科学基金(No.51674269),北京工业职业技术学院重点课题(Nos.bgzykyz201605,bgzyky201780z)

Mine Image Enhancement Algorithm Based on Dual Domain Decomposition

TIAN Zi-jian1, WANG Man-li1,2, WU Jun2, GUI Wei-feng2, WANG Wen-qing3   

  1. 1. School of Mechanical Electronic & Information Engineering, China University of Mining & Technology, Beijing 100083, China;
    2. School of Physics & Electronic Information Engineering, Henan Polytechnic University, Jiaozuo, Henan 454000, China;
    3. Beijing Polytechnic College, Beijing 100042, China
  • Received:2018-12-04 Online:2019-05-25 Published:2019-03-18
  • Supported by:

    National Natural Science Foundation of China (No. 51674269), Key Research Topics of Beijing Polytechnic College (Nos. bgzykyz201605, bgzyky201780z)

摘要:

为提高矿井下图像的对比度,并同步地抑制图像的雾尘和噪声,提出一种基于双域分解的矿井下图像增强算法.首先,采用双边滤波器将输入图像分解为低频图像和高频图像;其次,采用快速暗原色去雾算法和Gamma变换,实现低频图像的去雾和对比度提高;接着,采用非下采样Shearlet变换和二阶微分算子,实现高频图像降噪和增强;最后,将增强的低频、高频图像合成基础增强图像,并抑制粉尘散射模糊和过曝光白色伪影,得到最终增强图像.实验表明,该方法不仅能有效提高矿井下图像的对比度,还能有效抑制图像的雾气和噪声,具有广泛的应用前景.

关键词: 图像分解, 图像去雾, 图像降噪, 图像增强, 图像重构

Abstract:

To better the contrast of the images under the mine and suppress the influence of dust and noise, a mine image enhancement algorithm based on dual domain decomposition is proposed. Firstly, input images are decomposed into low frequency images and high frequency images respectively using bilateral filter. Secondly, haze removal and contrast enhancement of the low frequency images are realized by using the fast-dark channel prior dehazing algorithm and gamma transform. After that, non-subsampled Shearlet transform and second order differential operator are adopted to realize the high-frequency image denoising and enhancement. At last, the enhanced high-frequency and low-frequency images are composited into basic enhanced images followed by suppression of dust blur and over-exposure white artifact to obtain the desirable enhanced images. Result of the experiment shows that our algorithm is widely applicable for it not only can enhance the contrast of the images under the mine but also suppress the haze and noise effectively.

Key words: Image enhancement, Image decomposition, Image dehazing, Image reconstruction, Image denoising

中图分类号: