基于小波变换纹理一致性测度的遥感图像融合算法

Remote sensing image fusion algorithm using texture homogeneity measure based on wavelet transform

本文提出了一种基于小波变换的遥感图像融合算法,利用多分辨小波变换的系数,采用低频图像的小波系数最小值作为融合后的低频系数,高频图像根据纹理一致性测度的纹理检测确定融合规则,调整高频小波系数大小。利用小波变换对图像相对应的低频分量及各方向细节分量进行针对性融合处理,很好地将来自不同图像的特征与细节融合在一起,并对融合图像质量进行了对比评价。实验结果表明,这种方法能够在保留图像微小细节方面获得满意的结果,这种算法有效且优于传统的图像融合方法。

An image fusion algorithm of remote sen sing images based on wavelet transform is proposed. The approach processes multi-resolution wavelet coefficients individually and adopts minimum of low-frequency image wavelet coefficients as the fusion low-frequency image wavelet coefficients. High-frequency image fusion rules are based on the texture detection algorithm of texture homogeneity measure, and are used to adjust high-frequency image wavelet coefficients. Using wavelet transform, each low frequency component and orientation detail component of the image are fused pertinently. All the features and details from different images are fused successfully and the qualities of these fusion images are compared. Experimental results show that the proposed algorithm retains subtle image detail and achieves an excellent effect, and the new image fusion algorithm is much better than traditional ones.