计算机科学 ›› 2023, Vol. 50 ›› Issue (6): 209-215.doi: 10.11896/jsjkx.220300236
张历洪, 叶军
ZHANG Lihong, YE Jun
摘要: 高光谱图像(Hyperspectral Image,HSI)在采集过程中会产生多种类型的噪声,噪声数量越多,HSI的有效信息就越少。为了更有效地从大量混合噪声中恢复HSI的有效消息,文中提出了一种基于群稀疏正则化的约束平滑秩近似HSI恢复方法。其中,群稀疏正则化被定义为基于加权$\ell_{2,1}$范数的空谱全变分,该正则化在利用空谱维信息的同时也考虑到了HSI内部的群稀疏性,增强了模型对混合噪声的去除效果及空谱维的光滑性。此外,文中采用约束的平滑函数来近似秩函数,以更好地利用HSI的低秩属性并提高了算法效率。该优化问题采用基于交替方向乘子的迭代算法进行求解。两种加噪情况的模拟数据实验和一项基于真实数据的实验的结果表明,相比5种目前主流的方法,所提方法在目视效果和评价指标上都有明显提升。
中图分类号:
[1]XU M,SUN J,ZHOU X,et al.Research on nondestructive identification of grape varieties based on EEMD-DWT and hyperspectral image[J].Journal of Food Science,2021,86(5):2011-2023. [2]CHENG C,LI H,PENG J,et al.Hyperspectral image classification via spectral-spatial random patches network[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2021,14:4753-4764. [3]D ABOV,FOI A,KATKOVNIK V,et al.Image denoising by sparse 3-D transform-domain collaborative filtering[J].IEEE Transactions on Image Processing,2007,16(8):2080-2095. [4]WRIGHT J,GANESH A,RAO S,et al.Robust principal component analysis:Exact recovery of corrupted low-rank matrices via convex optimization[C]//Neural Information Processing Systems.2009:2080-2088. [5]ZHANG H,HE W,ZHANG L,et al.Hyperspectral image restoration using low-rank matrix recovery[J].IEEE Transactions on Geoscience & Remote Sensing,2014,52(8):4729-4743. [6]XIE Y,QU Y,TAO D,et al.Hyperspectral image restoration via iteratively regularized weighted schatten p-norm minimization[J].IEEE Transactions on Geoscience and Remote Sensing,2016,54(8):4642-4659. [7]YE H,LI H,YANG B,et al.A novel rank approximation me-thod for mixture noise removal of hyperspectral images[J].IEEE Transactions on Geoscience & Remote Sensing,2019,57(7):4457-4469. [8]ZHENG Y,HUANG T,ZHAO X,et al.Mixed noise removal in hyperspectral image via low-fibered-rank regularization[J].IEEE Transactions on Geoscience & Remote Sensing,2020,58(1):734-749. [9]ZHANG H,CAI J,HE W,et al.Double low-Rank matrix decomposition for hyperspectral image denoising and destriping[J].IEEE Transactions on Geoscience and Remote Sensing,2021,60:1-19. [10]ZHAO J,TIAN S,GEIS C,et al.Spectral-Spatial classification integrating band selection for hyperspectral imagery with severe noise bands[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2020,13:1597-1609. [11]HE W,ZHANG H Y,SHEN H F,et al.Hyperspectral image denoising using local low-rank matrix recovery and global spatial-spectral total variation[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2018,11(3):713-729. [12]YE J,ZHANG X.Hyperspectral image denoising via subspace low-rank representation and spatial-spectral total variation[J].Journal of Imaging Science and Technology,2020,64(1):10507-1-10507-9. [13]CHEN Y,HE W,YOKOYA N,et al.Hyperspectral image restoration using weighted group sparsity-regularized low-rank tensor decomposition[J].IEEE Transaction Cybernetics,2020,50(8):3556-3570. [14]ZHENG Y,HUANG T,ZHAO X,et al.Double factor regula-rized low-rank tensor factorization for mixed noise removal in hyperspectral image[J].IEEE Transactions on Geoscience & Remote Sensing,2020,58(12):8450-8464. [15]CHANG Y,YAN L,FANG H,et al.Anisotropic spectral-spatial total variation model for multispectral remote sensing image destriping[J].IEEE Transaction on Image Processing,2015,24(6):1852-1866. [16]ECKSTEIN J,YAO W.Understanding the convergence of thealternating direction method of multipliers:Theoretical and computational perspectives[J].Pacific Journal of Optimization,2015,11(4):619-644. [17]ATO P D,AN L.Convex analysis approach to D.C.Programming:Theory,algorithms and applications[J].Acta Mathema-tica Vietnamica,1997,22(1):289-356. [18] SUN L,JEON B,SOOMRO B,et al.Fast superpixel based subspace low rank learning method for hyperspectral denoising[J].IEEE Access,2018,6:12031-12043. [19] PENG J,XIE Q,ZHAO Q,et al.Enhanced 3DTV regularization and its applications on HSI denoising and compressed sensing[J].IEEE Transactions on Image Processing,2020,29:7889-7903. |
|