基于小波变换的红外图像融合算法研究

小波变换具有良好的多分辨分析特性,可用于可见光图像与红外图像的图像融合。一般情况下,对于分解后得到的低频分量采用加权平均的方法来进行简单处理,这对融合后图像的对比度和视觉效果有很大的影响。采用基于小波变换的图像融合算法来分别处理分解后的低频分量和高频分量,并与其他融合方法进行分析比较。实验结果表明,使用所提出的算法进行融合后的图像内容清晰,具有增强图像的空间细节能力,提高图像分辨效果和人眼对场景目标的发现和识别概率,在不影响互信息量值的情况下,融合后图像的边缘保持度提高了将近2倍。

Infrared Image Fusion Algorithm Based on Wavelet Transforms

Wavelet transform can be used in the fusion of visible image and infrared image due to its multi-resolution analytical characteristics.Generally,weighted average method is used to process the low-frequency components which will affect the contrast and the visible effect.In this paper,a new method based on wavelet transform is applied to process the low-frequency and high-frequency components separately,and is compared with other methods.Experimental results show that the fused image processed with the proposed method is more distinct and with more spatial detail information.The image resolution and target identification possibility are increased,and the of the fused image is nearly twice as the original image.