基于NSCT变换的图像融合及鲁棒性分析

Image Fusion Method Based on NSCT and Robustness Analysis

红外与可见光的图像融合技术可以有效提高图像的对比度和清晰度,增强夜视效果。非降采样的Contourlet变换在图像融合领域取得了一定的研究成果。提出一种基于区域标准差比例加权的非降采样Contourlet变换的图像融合方法,并对该方法的鲁棒性进行分析。首先对来自同一场景的配准后的红外与可见光图像进行非降采样Contourlet变换;其次对近似分量取平均,高频细节分量按照区域标准差比例加权求和;然后通过非降采样Contourlet反变换得到融合图像;最后通过大量实验,与Laplace变换、小波变换及Contourlet变换的结果进行比较,并通过噪声实验对各变换进行了鲁棒性分析。结果表明:非降采样Contourlet变换可以获得较好的融合效果,并且具有较高的鲁棒性。

Infrared and visible image fusion technology can effectively improve the image contrast and clarity, enhance night vision. Non-subsampled Contourlet transform （NSCT） in image fusion field has made some achievements. A regional standard deviation-weighted image fusion method based on non-subsampled Contourlet transform was proposed, and the robustness of the method was analyzed. Firstly, the registered infrared and visible images from the same scene were transformed by non-subsampled Contourlet transform, followed by approximate weight was averaging and weighted summing of high-frequency detail components in accordance with the proportion of the regional standard deviation and then the fusion image is obtained by inverse non-subsampled Contourlet transform, and then the fusion images were compared with the results obtained by Laplace transformation, wavelet transformation and Contourlet transformation through a large number of experiments, and the robustness analysis was done through the noise test. The results show that： non-subsampled Contourlet transform can achieve better fusion effect, and high robustness.