基于方向金字塔框架变换的遥感图像融合算法

为了综合利用多光谱遥感图像与全色遥感图像之间的互补信息，提出了一种方向金字塔框架变换(SPFT)，并基于此变换提出了一种遥感图像融合算法。具体融合过程是将多光谱图像的每个波段分别与高分辨力全色图像进行融合，首先将高分辨力全色图像与多光谱图像的待融合波段进行直方图匹配，然后对该波段图像以及直方图匹配后的高分辨力全色图像分别进行方向金字塔框架变换分解，融合过程就是对两图像方向金字塔框架变换分解后的系数进行组合，最后对组合后的系数进行方向金字塔框架逆变换即可得到该波段图像与高分辨力全色图像的融合图像。实验结果表明该算法在性能上优于基于亮度-色调-饱和度(1HS)的彩色空间变换以及基于离散小波框架变换(DWFT)的遥感图像融合方法，尤其对源图像之间存在配准误差的情况。

Remote Sensing Image Fusion Based on Steerable Pyramid Frame Transform

In order to integrate two registered high spatial resolution panchromatic image and low spatial resolution multispectral image of the same scene, a new remote sensing image fusion algorithm based on a novel steerable pyramid frame transform (SPFT) is proposed. Each band of the multispectral image is combined with the panchromatic image. In the individual band fusion, histogram matching is first performed on the panchromatic image to obtain a new high spatial resolution image with a uniform histogram as the band to be fused of the multispectral image. Then, the histogram matched panchromatic image and the band to be fused of the multispectral image are both decomposed by the SPFT. The SPFT coefficients are combined to produce the composite coefficients. Finally, the fused image of the panchromatic image and the band to be fused of the multispectral image is obtained by performing inverse SPFT. Experimental results show that the proposed image fusion algorithm outperforms the traditional methods based on the intensity-hue-saturation (IHS) transform and discrete wavelet frame transform (DWFT), especially in situations where the source images are not perfectly registered.