基于非下采样轮廓波的红外与可见光图像融合

Fusion of Infrared and Visible Images Based on NSCT

传统图像融合方法主要聚焦于图像细节信息整合,容易损失图像背景信息,本文借助于非下采样轮廓波的多尺度分解能力和模糊逻辑特性,提出一种红外与可见光图像融合的算法。首先,使用非下采样轮廓波获取图像高频成分和低频成分;其次,利用模糊逻辑规则整合低频成分,使用区域空间频率整合图像高频成分;最后,经过非下采样轮廓波逆变换得到融合图像。实验结果表明,与传统图像融合方法相比,本文算法能够较好地保留可见光图像的背景信息,同时凸显红外目标信息。

Traditional image fusion methods mainly focus on the integration of image details, which is easy to lose the background information. In this paper, by means of the multi-scale decomposition ability of non-subsampled contour waves and the characteristics of fuzzy logic, an algorithm of infrared and visible image fusion is proposed. Firstly, the high-frequency and low-frequency components of the image are obtained by using non-subsampled contour waves. Secondly, fuzzy rules were used to integrate low-frequency components, and regional clarity was used to integrate high-frequency components. Finally, the fused image is obtained by the inverse contourwave transform. Experimental results show that, compared with traditional image fusion methods, the proposed algorithm can retain the background information of visible images and highlight the infrared target information.