计算机与现代化

• 图像处理 • 上一篇    下一篇

一种面向交通场景的快速图像增强算法

  

  1. (1.中国航天科工集团第二研究院研究生院,北京100854;2.中国航天科工集团第二研究院,北京100854)
  • 收稿日期:2015-01-20 出版日期:2015-06-16 发布日期:2015-06-18
  • 作者简介:李灵杰(1988-),男,河南巩义人,中国航天科工集团第二研究院研究生院硕士研究生,研究方向:图像处理; 唐宁(1963-),男,湖南东安人,中国航天科工集团第二研究院研究员,硕士生导师,博士,研究方向:智能交通,图像处理。

A Fast Image Enhancement Algorithm for Traffic Scene

  1.  (1. Graduate School, Second Academy of China Aerospace Science and Industry Corporation, Beijing 100854, China;2. Second Academy of China Aerospace Science and Industry Corporation, Beijing 100854, China)
  • Received:2015-01-20 Online:2015-06-16 Published:2015-06-18

摘要: 为提高夜间图像的增强效果,针对具体的交通场景提出一种改进的图像增强算法。通过改进自适应对比度增强算法(ACE)中放大系数的自适应函数,算法能够有针对性地增强图像亮度较低和对比度较低的区域,防止图像过度增强,并且能够增强图像的细节,克服传统对比度增强算法的不足。实验结果表明,该算法增强效果显著,针对具体夜间交通场景图像处理速度快,自适应效果好,具有消除“振铃效应”、颜色保真和较好的保留并增强图像边缘和细节的能力。

关键词: 图像增强, 夜间图像, 对比度, 自适应, 交通场景

Abstract: In this paper, a night image enhancement method is proposed for intelligent traffic monitoring to improve visual effect of night image. By improving the adaptive function of the adaptive contrast enhancement (ACE) algorithm in the amplification factor, the algorithm can specifically enhance lowbrightness and lowcontrast areas, to prevent excessive image enhancement, and can enhance the details of images, overcoming the shortcomings of traditional contrast enhancement algorithms. Experimental results show that the algorithm enhancement effects are significant, it is of a fast speed while processing nighttime traffic scene image, good adaptive effect with the capacity of weakening “ringing effects”, color fidelity and better retaining and enhancing the image edges and details.

Key words: image enhancement, night image, contrast, adaptive, traffic scene

中图分类号: