摘要
在低照度环境下拍摄的可见光图像可视性较差,若将其与红外图像直接融合会导致融合结果清晰度不理想。针对这一问题,该文提出一种基于对比度增强与多尺度边缘保持分解的图像融合方法。首先,在融合之前采用基于导向滤波的自适应增强算法提高可见光图像中暗区内容的可视性。其次,通过一种尺度感知边缘保持滤波器对输入图像进行多尺度分解。再次,应用频率调谐滤波构造显著图。最后,利用由导向滤波生成的权重图重构融合图像。实验结果表明,所提方法不仅可以使细节信息更突出,而且还能够有效地抑制伪影。
The visibility of the visible images is not good under the poor lighting condition.If the visible and infrared images are fused directly,the resolution of the fused images is not ideal.In order to solve this problem,a modified infrared and visible image fusion approach based on contrast enhancement and multi-scale edge-preserving is proposed.Firstly,an adaptive enhancement method based on the guided filter is adopted to enhance the visibility of dark region content in the visible image.Input images are then decomposed with a scale-aware edge-preserving filter.Subsequently,saliency maps of infrared and visible images are calculated on the basis of frequency-tuned filtering.Finally,the fused images are reconstructed with the weighting maps.Experiments show that the proposed scheme can not only make the detail information more prominent,but also suppress the artifacts effectively.
作者
朱浩然
刘云清
张文颖
ZHU Haoran;LIU Yunqing;ZHANG Wenying(School of Electronics and Information Engineering, Changchun University of Science and Technology Changchun 130022, China;Photoelectric Engineering College, Changchun University of Science and Technology, Changchun 130022, China;Academy of Opto-electronics, Chinese Academy of Sciences, Beijing 100094, China)
出处
《电子与信息学报》
EI
CSCD
北大核心
2018年第6期1294-1300,共7页
Journal of Electronics & Information Technology
关键词
图像融合
对比度增强
多尺度边缘保持分解
导向滤波器
Image fusion
Contrast enhancement
Multi-scale edge-preserving decomposition
Guided filter
