基于双树复小波变换与引导滤波的红外与可见光图像融合算法

Infrared-and-Visible-Image Fusion Algorithm based on Dual-Tree Complex Wavelet Transform and Guided Filtering

针对传统的红外与可见光图像融合中多尺度融合方法存在图像细节丢失、边缘模糊以及对比度下降等不足,提出了基于双树复小波变换(DTCWT)和引导滤波器的红外与可见光图像融合方法。首先,利用DTCWT分解源图像,获取源图像的低频子带系数和高频子带系数;其次,对分解得到的低频系数基于区域能量加权融合,对高频系数先基于区域能量取大融合,再应用引导滤波进行细节增强;最后,通过DTCWT逆变换得到融合图像。实验结果表明,融合后的图像能够较好地保留细节信息,同时使边缘更加清晰,具有更好的视觉效果,且信息熵、互信息量、边缘相似度等客观评价指标相对传统方法均有不同程度的提高。

The traditional multi-scale fusion method of infrared-and-visible-image fusion has the disadvantages of image-detail loss, edge blur and contrast reduction. Therefore, an infrared-and-visibleimage fusion method based on DTCWT(dual-tree complex wavelet transform) and guided filter is proposed.Firstly, the DTCWT is used to decompose the source image and obtain the low frequency subband coefficient and the high frequency subband coefficient of the source image. Then based on the regional energy, the low-frequency coefficients obtained by the decomposition are weighted and fused, and firstly based on the regional energy, the high-frequency coefficients are fused, and then the guided filtering is applied to enhancing the detail. Finally, the fused image is obtained by inverse transform of DTCWT. The experimental results show that the fused image can retain the details better, and make the edges clearer and have better visual effects. In addition, the objective evaluation indexes such as information entropy, mutual information,and edge similarity are improved in varying degrees as compared with the traditional methods.