光谱变化中基于Schatten-0范数正则化的高光谱和多光谱图像融合

Hyperspectral and Multispectral Image Fusion Based on Schatten-0 Norm Regularization Account for Spectral Variability

高光谱和多光谱图像融合旨在获取同时具有高空间分辨率和高光谱分辨率的高质量图像。然而,针对光谱变化中的高光谱和多光谱图像融合问题,全变分正则化方法仅仅是在空间梯度域对图像局部特性信息进行建模,没有考虑高光谱图像光谱信息间的高阶相关性。针对上述问题,通过引入Schatten-0正则项,实现对光谱信息高阶相关性的建模,提出基于Schatten-0范数正则化的高光谱和多光谱图像融合方法。采用交替方向乘子法(Alternating Direction Method of Multipliers,ADMM)求解光谱变化中的融合问题。其中,Schatten-0正则项对应的子问题采用硬阈值迭代收缩算法求解。仿真实验验证了所提方法的可行性和有效性。可为更具有实际价值、更一般化的高光谱和多光谱图像融合应用提供理论与技术支撑。

Hyperspectral and multispectral image fusion aims to obtain high-quality images with both high spatial resolution and high spectral resolution.However,for hyperspectral and multispectral image fusion account for spectral variability,the total variation regulari-zation method only models the local property of the images in the spatial gradient domain,without considering the high-order correlation between the spectral information of the hyperspectral image.Targeting at the above problem,a Schatten-0 regularizer is introduced to re-alize the modeling of high order correlation of spectral information,and a hyperspectral and multispectral image fusion method based on Schatten-0 norm regularization is proposed.The alternating direction method of multipliers(ADMM)is used to solve the related fusion problem.The subproblem corresponding to the Schatten-0 regularizer is solved by using hard threshold iterative shrinkage algorithm.The simulation results verify the feasibility and effectiveness of the proposed method,which can provide theoretical and technical support for more practical and general hyperspectral and multispectral image fusion applications.