Hyperspectral Image Denoising via Framelet Transformation Based Three-Modal Tensor Nuclear Norm

Kong, Wenfeng; Song, Yifan; Liu, Jing

doi:10.3390/rs13193829

Cited by 7 publications

(5 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the low-rank tensor approximation lost small content information, Zeng et al [25] have added regularization to the low-rank tensor approximation item, which can improve the performance of remote sensing image restoration. Kong et al [26] have proposed a framelet-tensor nuclear norm model for hyperspectral image denoising that takes full advantage of the redundancy of the framelet transform and the low-rank nature of the framelet-based transformed tensor. Overall, Tensor decompositionbased algorithms have excessively high computational and time costs, due to the need for decomposition on high-dimensional tensors.…”

Section: Related Work 21 Traditional Methods Of Remote Sensing Image ...mentioning

confidence: 99%

Anisotropic Weighted Total Variation Feature Fusion Network for Remote Sensing Image Denoising

Tan

2022

Remote Sensing

View full text Add to dashboard Cite

Remote sensing images are widely applied in instance segmentation and objetive recognition; however, they often suffer from noise, influencing the performance of subsequent applications. Previous image denoising works have only obtained restored images without preserving detailed texture. To address this issue, we proposed a novel model for remote sensing image denoising, called the anisotropic weighted total variation feature fusion network (AWTVF2Net), consisting of four novel modules (WTV-Net, SOSB, AuEncoder, and FB). AWTVF2Net combines traditional total variation with a deep neural network, improving the denoising ability of the proposed approach. Our proposed method is evaluated by PSNR and SSIM metrics on three benchmark datasets (NWPU, PatternNet, UCL), and the experimental results show that AWTVF2Net can obtain 0.12∼19.39 dB/0.0237∼0.5362 higher on PSNR/SSIM values in the Gaussian noise removal and mixed noise removal tasks than State-of-The-Art (SoTA) algorithms. Meanwhile, our model can preserve more detailed texture features. The SSEQ, BLIINDS-II, and BRISQUE values of AWTVF2Net on the three real-world datasets (AVRIS Indian Pines, ROSIS University of Pavia, HYDICE Urban) are 3.94∼12.92 higher, 8.33∼27.5 higher, and 2.2∼5.55 lower than those of the compared methods, respectively. The proposed framework can guide subsequent remote sensing image applications, regarding the pre-processing of input images.

show abstract

Section: Related Work 21 Traditional Methods Of Remote Sensing Image ...mentioning

confidence: 99%

Anisotropic Weighted Total Variation Feature Fusion Network for Remote Sensing Image Denoising

Tan

2022

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…In the second stage, to solve problem (14), it could first be divided into several subproblems, and then we applied the idea of the alternating direction method of multipliers (ADMM) [59] to update each variable.…”

Section: Lrsmd-tnnmentioning

confidence: 99%

Spectral–Spatial Complementary Decision Fusion for Hyperspectral Anomaly Detection

Xiang

Song

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

Hyperspectral anomaly detection has become an important branch of remote–sensing image processing due to its important theoretical value and wide practical application prospects. However, some anomaly detection methods mainly exploit the spectral feature and do not make full use of spatial features, thus limiting the performance improvement of anomaly detection methods. Here, a novel hyperspectral anomaly detection method, called spectral–spatial complementary decision fusion, is proposed, which combines the spectral and spatial features of a hyperspectral image (HSI). In the spectral dimension, the three–dimensional Hessian matrix was first utilized to obtain three–directional feature images, in which the background pixels of the HSI were suppressed. Then, to more accurately separate the sparse matrix containing the anomaly targets in the three–directional feature images, low–rank and sparse matrix decomposition (LRSMD) with truncated nuclear norm (TNN) was adopted to obtain the sparse matrix. After that, the rough detection map was obtained from the sparse matrix through finding the Mahalanobis distance. In the spatial dimension, two–dimensional attribute filtering was employed to extract the spatial feature of HSI with a smooth background. The spatial weight image was subsequently obtained by fusing the spatial feature image. Finally, to combine the complementary advantages of each dimension, the final detection result was obtained by fusing all rough detection maps and the spatial weighting map. In the experiments, one synthetic dataset and three real–world datasets were used. The visual detection results, the three–dimensional receiver operating characteristic (3D ROC) curve, the corresponding two–dimensional ROC (2D ROC) curves, and the area under the 2D ROC curve (AUC) were utilized as evaluation indicators. Compared with nine state–of–the–art alternative methods, the experimental results demonstrate that the proposed method can achieve effective and excellent anomaly detection results.

show abstract

“…Through simulation analysis, it has been pointed out that the designed algorithm has significantly improved computational speed and the ability to maintain good edges and structure. Literature [17] for the traditional denoising method ignores the spatial dimension problem, the frame-based tensor fiber rank is proposed as a new representation of the tensor rank, and the frame-based three-mode tensor kernel paradigm is proposed as its convex relaxation, and the numerical analysis results of several experiments show that this model maintains a good denoising performance and also provides information about the spatial dimension. Literature [18] aims at solving the problem of high light image acquisition affected by hybrid noise and proposes an HSI denoising method using local low-rank matrix recovery and 0-gradient, which can simultaneously recognize the low-rank structure of HIS and hybrid noise.…”

Section: Introductionmentioning

confidence: 99%

Research on the denoising algorithm for hyperspectral images based on tensor decomposition and full variational constraints

Chen

2024

Applied Mathematics and Nonlinear Sciences

View full text Add to dashboard Cite

In this paper, based on tensor decomposition, SSTV regular constraints are combined with low-rank 3D tensor for image denoising and the effect of the algorithm is enhanced by the augmented Lagrangian method to construct a hyperspectral image denoising algorithm based on tensor decomposition and full variational constraints. After the algorithm design is completed, image restoration is performed based on the use of objective evaluation, standard mean square error, and peak signal-to-noise ratio to test the specific effect of the algorithm. 2 sets of experiments were designed and analyzed the sensitivity of the algorithm parameters. The test results show that for the penalty parameter μ = C max ( m , n ) σ \mu = C\max \left( {\sqrt {\rm{m}} ,\sqrt {\rm{n}} } \right)\sigma , the optimal results are achieved when C=8 and K values are 13-15. The PSNR index of this paper's algorithm is always greater than 45 when the noise intensity is 0.025-0.1, the highest is 58.817, and the lowest is 45.837. The DN value of the image denoised by the algorithm floats 0.012-0.085 on the basis of the original curve, which is less than 0.1. The performance of the algorithm decreases as the noise becomes more and more complex, and the noise intensity of 0.1 finally drops to 45.837, but the output image is still clear.

show abstract

Hyperspectral Image Denoising via Framelet Transformation Based Three-Modal Tensor Nuclear Norm

Cited by 7 publications

References 44 publications

Anisotropic Weighted Total Variation Feature Fusion Network for Remote Sensing Image Denoising

Anisotropic Weighted Total Variation Feature Fusion Network for Remote Sensing Image Denoising

Spectral–Spatial Complementary Decision Fusion for Hyperspectral Anomaly Detection

Research on the denoising algorithm for hyperspectral images based on tensor decomposition and full variational constraints

Contact Info

Product

Resources

About