一种自适应的小波方向对比度图像融合算法

对多传感器获得的图像序列进行图像融合,可以采用基于小波变换的多分辨率分析图像融合方法。首先,对两幅待融合图像进行小波变换,采用平均加权的方法来获得融合后的低频分量;采用一种基于图像对比度的自适应算法来获得融合后的高频分量。最后从最高分解层到最低分解层依次对得到的高频小波系数和该分解层的低频小波系数进行小波逆变换,最终得到具有原图像有用信息的融合图像。实验结果表明,这种算法可以很好地保留原图像的有用信息,是一种有效的图像融合算法.

An Adaptive Direction of Wavelet Contrast Algorithm of Image Fusion

Images sequences obtained from the multi-sensor can be fused by the image fusion method based on multi-resolution analysis of wavelet transform.First,two images to be fused take the wavelet transform,weighted average method can be used to obtain the low-frequency fused components;a contrast adaptive algorithm based on image fusion can be used to obtain the high-frequency fused components.Finally,the high-frequency wavelet coefficients and the low-frequency wavelet coefficients on the same decomposition level have the inverse wavelet transform in turn from the highest to the lowest decomposition level,the final fused image containing the useful information of the original image will be obtained.