Hyperspectral Image Denoising with a Combined Spatial and Spectral Weighted Hyperspectral Total Variation Model

被引:31
|
作者
Jiang, Cheng [1 ,4 ]
Zhang, Hongyan [2 ,4 ]
Zhang, Liangpei [2 ,4 ]
Shen, Huanfeng [3 ,4 ]
Yuan, Qiangqiang [1 ,4 ]
机构
[1] Wuhan Univ, Sch Geodesy & Geomat, 129 Luoyu Rd, Wuhan 430079, Peoples R China
[2] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, 129 Luoyu Rd, Wuhan 430079, Peoples R China
[3] Wuhan Univ, Sch Resources & Environm Sci, 129 Luoyu Rd, Wuhan 430079, Peoples R China
[4] Wuhan Univ, Collaborat Innovat Ctr Geospatial Technol, 129 Luoyu Rd, Wuhan 430079, Peoples R China
基金
中国国家自然科学基金;
关键词
ANISOTROPIC DIFFUSION; NOISE-REDUCTION; SPARSE REPRESENTATION; REGRESSION; ALGORITHM; REMOVAL;
D O I
10.1080/07038992.2016.1158094
中图分类号
TP7 [遥感技术];
学科分类号
081102 ; 0816 ; 081602 ; 083002 ; 1404 ;
摘要
Hyperspectral image (HSI) denoising is a prerequisite for many subsequent applications. For an HSI, the level and type of noise often vary with different bands and spatial positions, which make it difficult to effectively remove noise while preserving textures and edges. To alleviate this problem, we propose a new total-variation model. The main contribution of the proposed approach lies in that the adaptive regularization terms in both the spatial and the spectral dimensions are designed separately and then combined into a unified framework. The 2 separate regularization terms allow a better description of the intrinsic nature of the original HSI data and can simultaneously penalize the noise from both the spatial and spectral perspectives. The designed weights for the regularization terms are positively correlated with the magnitude of the noise intensity and negatively correlated with the signal variation; thus, the original signal can be accurately retained and the noise can be effectively suppressed. To efficiently process the HSI, which appears as a huge data cube, a new optimization algorithm based on the alternating direction method of multipliers (ADMM) procedure is proposed to solve the new model. Experiments using HYDICE and AVIRIS images were conducted to validate the effectiveness of the proposed method.Resume. Hyperspectrale l'image (HSI) debruitage est une condition prealable pour de nombreuses applications ulterieures. Pour un HSI, le niveau et le type de bruit varie souvent avec differents groupes et positions spatiales, ce qui rend difficile d'eliminer efficacement le bruit tout en preservant les textures et les bords. Pour pallier ce probleme, nous proposons un nouveau modele de variation totale. Les principales contributions de l'approche proposee mensonge dans la conception des termes de regularisation adaptative dans les deux dimensions spatiales et spectrales, et en les combinant dans un cadre unifie. Les deux termes de regularisation separes permettent une meilleure description de la nature intrinseque des donnees HSI original et peuvent penaliser simultanement le bruit a la fois des perspectives spatiales et spectrales. Les poids concus pour les termes de regularisation sont en correlation positive avec la grandeur de l'intensite du bruit et correlation negative avec la variation de signal; ainsi, le signal d'origine peut etre retenu avec precision et le bruit peut etre efficacement supprimee. Pour traiter efficacement le HSI, qui apparait comme un enorme cube de donnees, un nouvel algorithme d'optimisation base sur la methode de direction alternee de multiplicateurs << alternating direction method of multipliers >> (ADMM) procedure est proposee pour resoudre le nouveau modele. Des experiences utilisant des images AVIRIS et HYDICE et ont ete menees afin de valider l'efficacite de la methode proposee.
引用
收藏
页码:53 / 72
页数:20
相关论文
共 50 条
  • [41] Hyperspectral Image Denoising via Group Sparsity Regularized Hybrid Spatio-Spectral Total Variation
    Zhang, Pengdan
    Ning, Jifeng
    REMOTE SENSING, 2022, 14 (10)
  • [42] A Light-Weighted Spectral-Spatial Transformer Model for Hyperspectral Image Classification
    Arshad, Tahir
    Zhang, Junping
    IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2024, 17 : 12008 - 12019
  • [43] A total variation denoising algorithm for hyperspectral data
    Li Ting
    Chen Xiao-mei
    Xue Bo
    Li Qian-qian
    Ni Guo-qiang
    INFRARED, MILLIMETER WAVE, AND TERAHERTZ TECHNOLOGIES, 2010, 7854
  • [44] Hyperspectral image restoration via RPCA model based on spectral-spatial correlated total variation regularizer
    Gao, Junheng
    Wang, Hailin
    Peng, Jiangjun
    NEUROCOMPUTING, 2025, 619
  • [45] SMDS-Net: Model Guided Spectral-Spatial Network for Hyperspectral Image Denoising
    Xiong, Fengchao
    Zhou, Jun
    Tao, Shuyin
    Lu, Jianfeng
    Zhou, Jiantao
    Qian, Yuntao
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2022, 31 : 5469 - 5483
  • [46] Structure tensor total variation-regularized weighted nuclear norm minimization for hyperspectral image mixed denoising
    Wu, Zhaojun
    Wang, Qiang
    Jin, Jing
    Shen, Yi
    SIGNAL PROCESSING, 2017, 131 : 202 - 219
  • [47] Spectral-Spatial Weighted Sparse Regression for Hyperspectral Image Unmixing
    Zhang, Shaoquan
    Li, Jun
    Li, Heng-Chao
    Deng, Chengzhi
    Plaza, Antonio
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2018, 56 (06): : 3265 - 3276
  • [48] HYPERSPECTRAL IMAGE RESTORATION BASED ON TENSOR AVERAGE RANK AND SPATIAL-SPECTRAL TOTAL VARIATION
    Wang, Fengfeng
    Li, Jie
    Yuan, Qiangqiang
    2022 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2022), 2022, : 7571 - 7574
  • [49] Adaptive total variation-based spectral-spatial feature extraction of hyperspectral image
    Zhang, Guoyun
    Wang, Jinping
    Zhang, Xiaofei
    Fei, Hongyan
    Tu, Bing
    JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2018, 56 : 150 - 159
  • [50] A novel weighted total variation model for image denoising
    Li, Meng-Meng
    Li, Bing-Zhao
    IET IMAGE PROCESSING, 2021, 15 (12) : 2749 - 2760