江汉大学学报(自然科学版) ›› 2020, Vol. 48 ›› Issue (6): 84-90.doi: 10.16389/j.cnki.cn42-1737/n.2020.06.012

• 计算机与信息科学 • 上一篇    下一篇

基于MATLAB 的图像去雾技术研究

齐卉a,孙超*b,苏通b,马俊智b,朱勇杰b,丁建军c   

  1. 江汉大学 a. 人工智能学院,b. 智能制造学院,c. 爆炸与爆破技术研究院,湖北 武汉 430056
  • 出版日期:2020-12-28 发布日期:2020-12-18
  • 作者简介:齐卉(1992— ),女,助理实验师,硕士,研究方向:信息获取、分析与处理。
  • 基金资助:
    SUN Chao

Study on Image Defogging Technology Based on MATLAB

QI Huia,SUN Chao*b,SU Tongb,MA Junzhib,ZHU Yongjieb,DING Jianjunc   

  1. a. School of Artificial Intelligence,b. School of Intelligent Manufacturing,c. Institute of Explosion Science and Blasting Technology,Jianghan University,Wuhan 430056,Hubei,China
  • Online:2020-12-28 Published:2020-12-18
  • Contact: 孙超
  • Supported by:
    国家自然科学基金资助项目(61941502);湖北省自然科学基金资助项目(2019CFB205);湖北省教育厅科学研究计划指导性项目(B2019242);武汉市教育局市属高校产学研项目(CXY201803)

摘要: 为了对雾霾天气所采集的退化图像做清晰化处理,在MATLAB 平台下构建开发了一个图像去雾系统。该系统能通过全局直方图均衡化、局部直方图均衡化、多尺度Retinex、暗通道先验4 种经典去雾算法对雾天降质图像进行不同程度的复原,统计图像质量评价指标对比去雾效果。实验结果表明,经过不同去雾算法处理的图像的清晰度和对比度都有所提高,实际应用中可根据场景需要选择相应算法。该系统的实现为图像去雾技术的研究奠定基础,具有良好的可操作性。

关键词: MATLAB, 图像去雾系统, 图像质量评价

Abstract: In order to restore the degraded images collected by outdoor acquisition devices in haze weather,this paper proposes an image defogging system on the MATLAB platform. The system can restore foggy degradation images to different degrees through four classic defogging algorithms:global histogram equalization,local histogram equalization,multiscale Retinex,and dark channel prior. The defogging effects are compared by image quality evaluation indicators. Experimental results show that the sharpness and contrast of the images processed by different defogging algorithms have been improved, and the corresponding algorithm can be selected according to the needs of the scene in practical applications. The implement of the system lays a foundation for the research of image defogging technology,and it has good operability.

Key words: MATLAB, image defogging system, image quality evaluation

中图分类号: