红外技术, 2017, 39 (4): 358, 网络出版: 2017-06-02   

边缘和对比度增强的 NSST域红外与可见光图像融合

Infrared and Visible Image Fusion Based on Improved Saliency Map in NSST Domain
作者单位
1 空军工程大学航空航天工程学院,陕西西安 710038
2 中国人民解放军 93787部队,北京 100076
摘要
为了将红外图像的全局信息与可见光图像的细节信息进行有效结合,进一步提高融合后图像的质量,提出了一种同时增强图像边缘细节和对比度的非下采样剪切波变换( NSST)域红外和可见光图像融合方法。首先,通过平移不变剪切波将图像分解成为低频子带与高频子带,通过全局显著性图分析图像的对比度信息;利用改进型局部显著度图分析图像局部边缘信息。针对不同频带系数,结合边缘信息和对比度信息对频带系数进行融合,最后,利用逆变换得到最终的融合图像。大量实验结果表明,本文方法在提高图像整体对比度的同时增强了图像的边缘细节表现能力,优于现有的基于小波变换,非下采样轮廓波变换(NSCT)和显著度图等几种图像融合方法。
Abstract
For the effective combination of global information of infrared image with detail information of visible image, and further improvement in the quality of the fused image, a fusion method of infrared and visible image based on enhancing contrast and edge of image in non-subsampled Shearlet transform(NSST) domain is proposed. Firstly, the image is decomposed to get the low frequency coefficients and the high frequency coefficients, and the image’s information of contrast is analyzed by global saliency map; the image’s information of edge is analyzed by improved local saliency map. For the different bands, the coefficients are fused using the information of contrast and edge, and the fused image is reconstructed by inverse NSST. A large number of experimental results show that the proposed method enhance the details in the fused image, while preserving the contrast of the image. And it is superior to the existing fused methods such as the method based on wavelet transform, non-subsampled Contourlet transform(NSCT), and the on saliency map and so on.

吴冬鹏, 毕笃彦, 马时平, 何林远, 张跃. 边缘和对比度增强的 NSST域红外与可见光图像融合[J]. 红外技术, 2017, 39(4): 358. WU Dongpeng, BI Duyan, MA Shiping, HE Linyuan, ZHANG Yue. Infrared and Visible Image Fusion Based on Improved Saliency Map in NSST Domain[J]. Infrared Technology, 2017, 39(4): 358.

本文已被 3 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!