A Convex Optimization-Based Coupled Nonnegative Matrix Factorization Algorithm for Hyperspectral and Multispectral Data Fusion

Lin, Chia-Hsiang; Ma, Fanyuan; Chi, Chong-Yung; Hsieh, Chih-Hsiang

doi:10.1109/tgrs.2017.2766080

Cited by 96 publications

(80 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where g ∈ R r 2 is a Gaussian kernel vector. In this article, F and G can be estimated from the datasets for the fusion method [25,32]. From Equations 2and 3, the objective function of original CNMF proposed in [25] is formulated as two coupled data fidelity terms; i.e., min…”

Section: Piecewise Spectral Smoothingmentioning

confidence: 99%

“…To lower the effect of data dimension reduction, spatial signature regularizers can be incorporated into the original fusion method, such as spatial smoothing [26,27] and sparse coding [28,29]. However, the sparsity-promoting regularizer itself may not be sufficient to yield high-quality fused data [28,30,31]; meanwhile, joint spatial and spectral regularization can perform well [32,33]. Charis et al proposed an NMF-based fusion method with several physical constraints by jointly unmixing the HIS and MSI data into pure reflectance spectra of the observed materials for hyperspectral super-resolution.…”

Section: Introductionmentioning

confidence: 99%

“…A regularized CNMF-based method was proposed [34] by introducing the volume of signature vectors' simplex regularizer, yet the heuristic algorithm did not improve the performance substantially. Lin et al [32] proposed a convex optimization-based CNMF based on the sum-of-squared distances (SSD) between all the endmembers, in which the sparsity and SSD-based regularizers were employed to bring significant improvements in fusion performance. However, this algorithm was not suitable for a high noise environment, yielding poor anti-noise performance.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Joint Spatial-Spectral Smoothing in a Minimum-Volume Simplex for Hyperspectral Image Super-Resolution

Yang

Ping

et al. 2019

Applied Sciences

Self Cite

View full text Add to dashboard Cite

The limitations of hyperspectral sensors usually lead to coarse spatial resolution of acquired images. A well-known fusion method called coupled non-negative matrix factorization (CNMF) often amounts to an ill-posed inverse problem with poor anti-noise performance. Moreover, from the perspective of matrix decomposition, the matrixing of remotely-sensed cubic data results in the loss of data’s structural information, which causes the performance degradation of reconstructed images. In addition to three-dimensional tensor-based fusion methods, Craig’s minimum-volume belief in hyperspectral unmixing can also be utilized to restore the data structure information for hyperspectral image super-resolution. To address the above difficulties simultaneously, this article incorporates the regularization of joint spatial-spectral smoothing in a minimum-volume simplex, and spatial sparsity—into the original CNMF, to redefine a bi-convex problem. After the convexification of the regularizers, the alternating optimization is utilized to decouple the regularized problem into two convex subproblems, which are then reformulated by separately vectorizing the variables via vector-matrix operators. The alternating direction method of multipliers is employed to split the variables and yield the closed-form solutions. In addition, in order to solve the bottleneck of high computational burden, especially when the size of the problem is large, complexity reduction is conducted to simplify the solutions with constructed matrices and tensor operators. Experimental results illustrate that the proposed algorithm outperforms state-of-the-art fusion methods, which verifies the validity of the new fusion approach in this article.

show abstract

Section: Piecewise Spectral Smoothingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Joint Spatial-Spectral Smoothing in a Minimum-Volume Simplex for Hyperspectral Image Super-Resolution

Yang

Ping

et al. 2019

Applied Sciences

Self Cite

View full text Add to dashboard Cite

show abstract

“…It is a well-known trick that the image resolution can be improved by learning the residual of the high-frequency component, instead of directly learn the SR result in RGB domain [10]. Also, there is a recent research trend in the remote sensing area (see, e.g., [17]), aiming at fusing the high-resolution spatial details (probably extracted from some counterpart panchromatic image) with an input lowresolution image. To take the advantages from different super-resolved results (or information from LR image), we need to have different SR models first.…”

Section: Dual High-resolution Image Reconstructionmentioning

confidence: 99%

“…This kind of image fusion approach is of paramount importance in nowadays satellite remote sensing technology, and also appears in other domain applications, such as the hyperspectral super-resolution technique [17] which fuses a high-resolution multispectral image with a low-resolution hyperspectral image in order to obtain a high spatial/spectral-resolution imagery-critical in military surveillance. However, these approaches all rely on high-resolution counterpart imageries that are, however, not always available, motivating us to pursue single-image super-resolution (SISR) for facilitating the ensuing computer vision tasks [8].…”

Section: Introductionmentioning

confidence: 99%

Dual Reconstruction with Densely Connected Residual Network for Single Image Super-Resolution

Hsu

Lin

2019

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

Self Cite

View full text Add to dashboard Cite

Deep learning-based single image super-resolution enables very fast and high-visual-quality reconstruction. Recently, an enhanced super-resolution based on generative adversarial network (ESRGAN) has achieved excellent performance in terms of both qualitative and quantitative quality of the reconstructed high-resolution image. In this paper, we propose to add one more shortcut between two dense-blocks, as well as add shortcut between two convolution layers inside a dense-block. With this simple strategy of adding more shortcuts in the proposed network, it enables a faster learning process as the gradient information can be back-propagated more easily. Based on the improved ESRGAN, the dual reconstruction is proposed to learn different aspects of the super-resolved image for judiciously enhancing the quality of the reconstructed image. In practice, the super-resolution model is pre-trained solely based on pixel distance, followed by fine-tuning the parameters in the model based on adversarial loss and perceptual loss. Finally, we fuse two different models by weighted-summing their parameters to obtain the final super-resolution model. Experimental results demonstrated that the proposed method achieves excellent performance in the real-world image super-resolution challenge. We have also verified that the proposed dual reconstruction does further improve the quality of the reconstructed image in terms of both PSNR and SSIM. Figure 1. Visual quality comparison between the (a) ground truth and the super-resolved results by using (b) Bicubic interpolation, (c) ESRGAN [26], and (d) proposed method.

show abstract

Multispectral Image Super-Resolution Using Structure-Guided RGB Image Fusion

Pan

Shen

2018

Pattern Recognition and Computer Vision

View full text Add to dashboard Cite

A Convex Optimization-Based Coupled Nonnegative Matrix Factorization Algorithm for Hyperspectral and Multispectral Data Fusion

Cited by 96 publications

References 54 publications

Joint Spatial-Spectral Smoothing in a Minimum-Volume Simplex for Hyperspectral Image Super-Resolution

Joint Spatial-Spectral Smoothing in a Minimum-Volume Simplex for Hyperspectral Image Super-Resolution

Dual Reconstruction with Densely Connected Residual Network for Single Image Super-Resolution

Multispectral Image Super-Resolution Using Structure-Guided RGB Image Fusion

Contact Info

Product

Resources

About