一种新的红外与可见光图像像素级融合方法

Novel Method on Pixel-level Fusion of Infrared and Visible Images

摘要:

提出了一种使用指导性滤波进行像素级图像融合的方法。首先对红外图像进行最大值滤波处理以获取其中的高亮信息;然后使用原始红外图像作为指导图像对高亮部分做滤波,处理结果直接叠加在可见光图像上;最后将叠加图像进行伽玛校正以调整对比度,得到最终融合输出。实验结果表明:该算法能够充分利用红外与可见光图像各自的优势,融合效果较好,且计算复杂度低,易于实时处理。

Abstract:

A pixel-level image fusion method using guided filter was proposed. Firstly, the infrared image was processed by means of maximum filter to obtain highlight information. Secondly, the original infrared image was used as a guidance image to modify the highlight parts, and the result was fused with the visible image directly. Finally, Gamma correction was done to adjust the contrast of the fused image. Experimental results indicate that the proposed algorithm can make good use of the advantages of infrared and visible images, and the sound fusion effect and real-time processing was achieved with low computational complexity.