为了从不同源图像中获取更多的互补信息,提出了一种基于提升机制的图像融合方法。在分析图像融合近几年研究成果的基础上,对融合图像进行双正交小波变换,获得不同尺度下的小波金字塔图像,再对低频系数取加权平均值,高频系数进行局部梯度运算,采用局部梯度值最大值作为高频系数,最后对处理的小波域的高低频系数反变换得到融合后图像。通过分析实验结果表明,提出的算法适合两幅及多幅图像的自主融合,而且可以获得视觉效果更好、细节更为丰富的融合图像,消除块效应。
In order to get more complementary information, an image fusion method based on lifting scheme is introduced. Analyzing the recent research results on image fusion, the biorthogonal wavelet transformation is taken on fusion image, and a different scale of wavelet pyramid image is obtained. After taking a weighted average value for the low frequency coefficients and the local gradient computation values for the high frequency coefficient, the fusion image is gotten by taking the inverse wavelet transform on the high and low frequency coefficients processed. By experimental analysis, the proposed algorithm is automatic and suitable for two and more images, and can give us better visual effects, more extensive integration of image detail while eliminating block effects.