熊承义,张梦杰, 高志荣, 龚忠毅.NSCT域梯度加权的红外与可见光图像融合[J].中南民族大学学报自然科学版,2018,(2):74-79
NSCT域梯度加权的红外与可见光图像融合
Fusion of Infrared and Visible Images Based on Weighted Gradient in NSCT Domain
  
DOI:10.12130/znmdzk.20180116
中文关键词: 图像融合  可见光图像  红外图像  非下采样轮廓波变换  梯度加权
英文关键词: image fusion  visible images  infrared images  NSCT  weighted gradient
基金项目:国家自然科学基金资助项目( 61471400)
作者单位
熊承义1,张梦杰1, 高志荣2, 龚忠毅1 1 中南民族大学 电子信息工程学院武汉 430074
2 中南民族大学 计算机科学学院武汉 430074 
摘要点击次数: 777
全文下载次数: 479
中文摘要:
      提出了一种非下采样轮廓波变换( NSCT) 域梯度加权的红外与可见光图像融合方法.在多尺度变换图像融合框架下,先对多源传感图像进行非下采样轮廓波变换,然后对变换得到的图像低通成分进行梯度域加权融合, 对高通成分进行绝对值最大选择融合,最后通过非下采样轮廓波逆变换得到融合图像.利用梯度加权融合生成融合 图像的低通系数,更好地保留了各源图像低通成分包含的有用信息.大量实验结果表明: 该方法能有效提升图像融合性能,明显增强融合图像的对比度,融合图像表现出更好的视觉质量和可观测性.
英文摘要:
      A fusion method of infrared and visible images based on weighted gradient in NSCT ( WGN) domain is proposed in this paper. The two source images are firstly decomposed into the corresponding low-pass components and the high-pass ones by non-subsampled contourlet transform, which is under the multi-scale transform based fusion framework. Then, the low-pass components are fused by the proposed weighted gradient fusion rule, and the high-pass components are fused by the rule of selecting maximal absolute values of coefficients. Finally, the fused image is obtained by performing the inverse NSCT on the merged coefficients. The low-pass coefficients fused by the weighted gradient method preserve more desirable information from source images. Large amounts of experimental results demonstrate that this fusion method can effectively improve the performance for the fusion of images with markedly enhancing the contrast of the fused image, and thus the fused image has better image quality and observability.
查看全文   查看/发表评论  下载PDF阅读器
关闭