作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程

• 图形图像处理 • 上一篇    下一篇

基于视觉感知的低照度图像增强算法

王小元a,b,张红英a,b,吴亚东c,刘言a,b   

  1. (西南科技大学 a.信息工程学院; b.特殊环境机器人技术四川省重点实验室; c.计算机科学与技术学院,四川 绵阳 621010)
  • 收稿日期:2015-07-06 出版日期:2016-08-15 发布日期:2016-08-15
  • 作者简介:王小元(1989-),女,硕士研究生,主研方向为图形图像处理;张红英、吴亚东,教授、博士;刘言,硕士研究生。
  • 基金资助:
    中国科学院“西部之光”人才培养计划基金资助项目(13zs0106);四川省科技厅科技支撑计划基金资助项目(2014SZ0223,2015GZ0212);特殊环境机器人技术四川省重点实验室开放基金资助项目(13zxtk05)。

Low-illumination Image Enhancement Algorithm Based on Visual Perception

WANG Xiaoyuan a,b,ZHANG Hongying a,b,WU Yadong c,LIU Yan a,b   

  1. (a.School of Information Engineering; b.Robot Technology Used for Special Environment Key Laboratory of Sichuan Province; c.School of Computer Science and Technology, Southwest University of Science and Technology,Mianyang,Sichuan 621010,China)
  • Received:2015-07-06 Online:2016-08-15 Published:2016-08-15

摘要: 针对低照度彩色图像的低亮度和低对比度特点,通过研究瞳孔及感光细胞对环境的自动调节过程,提出一种基于视觉感知的自适应亮度增强算法。效仿瞳孔对环境变化的适应过程,提升图像整体亮度。模拟感光细胞对低照度环境的自适应调控能力,设计暗适应函数与明适应函数。根据光照分布情况确定明暗信息融合函数,实现亮度分量的全局自适应调节。采用指数函数对邻域内像素点进行调整,提高亮度图像的局部对比度并对增强图像进行色彩还原。实验结果表明,该算法可有效增强低照度彩色图像中暗区及高光区的细节表现力,提高图像分析识别系统、视频监控系统等计算机视觉系统在低照度环境下的工作效率。

关键词: 低照度, 人眼视觉, 明适应, 暗适应, 对比度增强

Abstract: Aiming at the low brightness and low contrast of the low-illumination color image,an adaptive brightness enhancement algorithm based on visual perception is proposed by researching the automatic adjustment of the pupil and the photoreceptor cells to the environment.The overall brightness of the image is enhanced by simulating the adaptation of the pupil to the varieties of environments.The dark and light adaptive functions are designed by simulating the adaptive adjustment of photoreceptor cells for the low illumination environment.The fusion function is determined according to light distribution to adjust the global luminance adaptively.The exponential function is used in the neighborhoods to improve local contrast of the luminance component.The enhanced image is obtained by making color rendition on the enhanced luminance component.Experimental result show that the proposed algorithm can effectively improve the performance of details in dark and highlight areas.It can improve the working efficiency of the image analysis and recognition system,video surveillance system and other computer vision systems in low-illumination environment effectively.

Key words: low-illumination, human visual, light adaptation, dark adaptation, contrast enhancement

中图分类号: