系统工程与电子技术 ›› 2023, Vol. 45 ›› Issue (8): 2395-2404.doi: 10.12305/j.issn.1001-506X.2023.08.13

• 电子技术 • 上一篇    下一篇

基于图像特征的红外与低照度图像融合

王慧赢1, 王春平1, 付强1,*, 韩子硕2, 张冬冬1   

  1. 1. 陆军工程大学石家庄校区电子与光学工程系, 河北 石家庄 050003
    2. 中国人民解放军32356部队, 青海 西宁 710003
  • 收稿日期:2022-03-25 出版日期:2023-07-25 发布日期:2023-08-03
  • 通讯作者: 付强
  • 作者简介:王慧赢(1992—), 女, 博士研究生, 主要研究方向为图像处理、计算机视觉
    王春平(1965—), 男, 教授, 博士, 主要研究方向为图像处理、火力控制理论与应用
    付强(1981—), 男, 讲师, 博士, 主要研究方向为智能视觉与目标检测
    韩子硕(1986—), 男, 博士研究生, 主要研究方向为图像处理、计算机视觉
    张冬冬(1993—), 男, 硕士研究生, 主要研究方向为目标识别
  • 基金资助:
    军内科研项目(LJ20191A040155)

Infrared and low illumination image fusion based on image features

Huiying WANG1, Chunping WANG1, Qiang FU1,*, Zishuo HAN2, Dongdong ZHANG1   

  1. 1. Department of Electronic and Optical Engineering, Shijiazhuang Campus, Army Engineering University of PLA, Shijiazhuang 050003, China
    2. Unit 32356 of PLA, Xining 710003, China
  • Received:2022-03-25 Online:2023-07-25 Published:2023-08-03
  • Contact: Qiang FU

摘要:

针对难以独立分析红外图像和低照度图像场景中目标和背景信息的问题, 提出了基于图像特征的红外与低照度图像融合算法。首先, 针对性地对红外和低照度图像进行图像处理, 优化各自特性。其次, 采用非下采样剪切波变换对两者进行高低频分解。然后, 采用改进的拉普拉斯加权算法融合低频图像, 采用改进的脉冲耦合神经网络融合高频图像。最后, 经非下采样剪切波逆变换获取融合图像。实验结果表明, 所提算法可有效融合红外与低照度图像, 降低了噪声对融合图像的影响, 提高了融合图像的清晰度, 且图像对比度适中, 符合人眼视觉感知效果。

关键词: 低照度图像, 红外图像, 非下采样剪切波变换, 脉冲耦合神经网络, 图像融合算法

Abstract:

Aiming at the problem that it is difficult to analyze the target and background information in the scene of infrared images and low illumination images independently, an infrared and low illumination image fusion algorithm based on image features is proposed. Firstly, the infrared and low illumination images are processed to optimize their characteristics respectively. Secondly, the non-subsampled shearlet transform is used to decompose them in high and low frequency. Thirdly, the improved Laplace weighted algorithm is used to fuse the low-frequency images. The improved pulse coupled neural network is used to fuse the high-frequency images. Finally, the fused image is obtained by non-subsampled shearlet inverse transform. Experimental results show that the proposed algorithm can effectively fuse infrared and low illumination images, reduce the impact of noise on the fused image, improve the clarity of the fused image, and the contrast of the image is moderate, which is suifable the visual perception effect of human eyes.

Key words: low illumination image, infrared image, non-subsampled shearlet transform, pulse-coupled neural network, image fusion algorithm

中图分类号: