为了提高全色图像与多光谱图像的融合质量,提出一种基于非下采样双树复轮廓波变换和稀疏表示的图像融合算法.对多光谱图像进行亮度-色度-饱和度变换,再对亮度分量和全色图像进行直方图匹配及亮度平滑滤波处理.利用非下采样双树复轮廓波变换处理亮度分量和全色图像,得到对应的高低频系数.对于低频系数,利用稀疏表示进行融合,采用空间频率和l1范数双指标取大的融合规则得到稀疏表示系数;对于高频系数,将改进的拉普拉斯能量和作为脉冲耦合神经网络的外部输入项,提出了改进的脉冲耦合神经网络的融合策略.最后进行非下采样双树复轮廓波逆变换和亮度-色度-饱和度逆变换得到融合结果.实验结果表明,该算法最大限度地保留光谱信息的同时可以提高空间分辨率,视觉效果及客观指标均优于经典的融合算法.

In order to improve the fusion quality of multispectral image and panchromatic image, a remote sensing image fusion algorithm was proposed based on Non-subsampled Dual-tree Complex Contourlet Transform(NSDTCT) and sparse representation. Firstly, the Intensity-Hue-Saturation(IHS) transform was applied to the multispectral image. Then, the histogram matching and smoothing filter-based intensity modulation were used to handle intensity component and panchromatic image. Secondly, the NSDTCT was employed to decompose the new intensity component and panchromatic image, and the low frequency coefficients and high frequency coefficients were obtained. For the low frequency coefficients, a fusion method based on sparse representation was presented, and the fused coefficients were obtained by combining spatial frequency with l1-norm maximum. For the high frequency coefficients, the sum-modified Laplacian was used for the external input of Pulse Coupled Neural Network(PCNN), and a fusion method based on the theory of improved PCNN was presented. Finally, the fused image was obtained by inverse NSDTCT and inverse IHS transform. The experimental results show that the proposed algorithm can improve the spatial resolution and maintain the spectral characteristics simultaneously, and outperforms other classical fusion algorithms in terms of both the visual quality and objective evaluation.