基于双密度双树复数小波变换的图像融合研究

双密度双树复数小波变换综合了双密度小波、双树小波和复数小波的优点,具有平移不变性以及良好的方向性,将其引入图像融合能够更好地提取原始图像的特征,为融合图像提供更多的信息.该文提出一种基于双密度双树复数小波变换的图像融合方法,首先将图像进行双密度双树复数小波变换,然后在相应尺度上利用融合规则将变换系数融合,最后进行重构得到融合结果.采用多组具有不同特征的源图像进行融合实验,并对融合图像进行了主客观评价.实验结果表明,该文方法的融合效果优于传统的小波变换方法.

Research on Image Fusion with Double Density Dual-Tree Complex Wavelet Transform

Double Density Dual-Tree Complex Wavelet Transform (DDDTCWT) integrates the advantages of double density wavelet, dual tree wavelet and complex wavelet. It has translation invariability and improved directionality. When it is introduced to image fusion, the characteristics of original images are extracted better and more information for fusion is obtained. A novel image fusion method based on DDDTCWT is proposed in this paper. Firstly, the source images are decomposed using DDDTCWT, then the transform coefficients are fused with the fusion regular in the corresponding scales, and finally the fused coefficients are reconstructed to obtain fusion results. The proposed method is successfully used to merge several sets of multi-sensor images with different modalities. The experimental results indicate that the proposed approach can significantly outperform the traditional wavelet-transform-based image fusion method.