• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2015, Vol. 37 ›› Issue (09): 1750-1755.

• 论文 • 上一篇    下一篇

车载红外夜视图像的小波-遗传图像增强算法

于洋,康雪雪   

  1. (沈阳理工大学信息科学与工程学院,辽宁 沈阳 110159)
  • 收稿日期:2014-07-03 修回日期:2014-08-14 出版日期:2015-09-25 发布日期:2015-09-25
  • 基金资助:

    辽宁省教育厅资助项目(LT2012005)

A waveletgenetic enhancement algorithm
for vehicular infrared night vision images  

YU Yang,KANG Xuexue   

  1. (School of Information Science and Engineering,Shenyang Ligong University,Shenyang 110159,China)
  • Received:2014-07-03 Revised:2014-08-14 Online:2015-09-25 Published:2015-09-25

摘要:

针对传统红外图像增强算法在视觉效果上不够理想的问题,提出了一种适用于车载红外夜视图像的图像增强方法。该方法利用红外夜视仪的非接触生成热图像的原理,建立了针对车载红外夜视图像的小波遗传灰度图像增强方法,并将该方法与传统的直方图均衡化法和多尺度Retinex算法进行了对比。在红外夜视图像增强效果方面,该方法具有改善图像亮度均匀性、避免图像过分增强和抑制噪声的特点。实验表明,所提出的小波遗传图像增强算法在车载红外夜视图像增强方面的处理效果较好。

关键词: 红外夜视图像, 小波遗传, 遗传算法, 图像增强

Abstract:

Due to the fact that the traditional infrared image enhancement methods have imperfect visual effect, we propose an image enhancement method for vehicular infrared night vision images. We establish a waveletgenetic grayscale image enhancement algorithm on the basis of the noncontact thermal image generation principle of the infrared nightvision goggles. Compared with the traditional histogram equalization method and the multiscale Retinex algorithm, the proposed method can improve the image brightness uniformity, avoid excessive enhancement and effectively suppress noise. Experimental results show that the waveletgenetic enhancement algorithm can achieve good performance in enhancing vehicular infrared night vision images.

Key words: infrared night vision image;wavelet-genetic;genetic algorithm;image enhancement