基于导向滤波和小波变换的红外可见光图像融合改进算法研究

Research on Improved Infrared Visible Light Image Fusion Algorithm Based on Guided Filtering and Wavelet Transform

针对传统模型在跨模态下易产生光晕伪影、颜色失真等问题,提出一种基于导向滤波和小波变换的红外可见光图像融合改进算法。将源图像经由小波变换获得二维低频及高频的子代系数,低频分量采用加权平均融合,高频分量提取权重图后经导向滤波获得细节增强;再将所处理的各分量经小波逆变换获得融合图像。该算法使用开源数据集TNO检验效果,经过主客观评估,得出该算法的效果明显优于传统算法,符合研究预期。

Aiming at the problem that the traditional model is prone to produce halo artifacts and color distortion in cross-mode,an improved infrared visible light image fusion algorithm based on guided filtering and wavelet transform is proposed.The sub-generation coefficients of two-dimensional low frequency and high frequency are obtained from the source image through wavelet transform.The low frequency components are fused by weighted average,and the high frequency components are extracted from the weight map,and then the details are enhanced by guided filtering;then the processed components are transformed by inverse wavelet transform to obtain the fused image.The algorithm uses the open source data set TNO to test the effect.After subjective and objective evaluation,it is concluded that the effect of the algorithm is significantly better than the traditional algorithm,which is in line with the research expectation.