Compressive sensing and adaptive direct sampling in hyperspectral imaging

Hahn, Jürgen T.; Debes, Christian; Leigsnering, Michael; Zoubir, Abdelhak M.

doi:10.1016/j.dsp.2013.12.001

Cited by 57 publications

(27 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multispectral images are considered the holy grail of observation tools in Earth observation because they can provide both spectral and spatial information of an object or scene. A multi-spectral camera is usually used as a payload of a satellite in multiple applications, such as monitor environments, surveying minerals, military targets, and so on [6][7][8][9]. Unfortunately, the information content of a multispectral image with large spatial and spectral information (high spectral and spatial resolution) [10,11] is much greater than the tolerance capabilities of current on-orbit available memory [12] and image-transmission downlink bandwidth of satellites.…”

Section: Introductionmentioning

confidence: 99%

Multispectral Transforms Using Convolution Neural Networks for Remote Sensing Multispectral Image Compression

Liu

2019

Remote Sensing

View full text Add to dashboard Cite

A multispectral image is a three-order tensor since it is a three-dimensional matrix, i.e.one spectral dimension and two spatial position dimensions. Multispectral image compression canbe achieved by means of the advantages of tensor decomposition (TD), such as NonnegativeTucker Decomposition (NTD). Unfortunately, the TD suffers from high calculation complexity andcannot be used in the on-board low-complexity case (e.g., multispectral cameras) that the hardwareresources and power are limited. Here, we propose a low-complexity compression approach formultispectral images based on convolution neural networks (CNNs) with NTD. We construct anew spectral transform using CNNs, where the CNNs are able to transform the three-dimensionspectral tensor from large-scale to a small-scale version. The NTD resources only allocate thesmall-scale three-dimension tensor to improve calculation efficiency. We obtain the optimizedsmall-scale spectral tensor by the minimization of original and reconstructed three-dimensionspectral tensor in self-learning CNNs. Then, the NTD is applied to the optimized three-dimensionspectral tensor in the DCT domain to obtain the high compression performance. We experimentallyconfirmed the proposed method on multispectral images. Compared to the case that the newspectral tensor transform with CNNs is not applied to the original three-dimension spectral tensorat the same compression bit-rates, the reconstructed image quality could be improved. Comparedwith the full NTD-based method, the computation efficiency was obviously improved with only asmall sacrifices of PSNR without affecting the quality of images.

show abstract

Section: Introductionmentioning

confidence: 99%

Multispectral Transforms Using Convolution Neural Networks for Remote Sensing Multispectral Image Compression

Liu

2019

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…In general, the measurement matrix is required to be incoherent with the original HSI, where Gauss random matrix as the measurement matrix is a better choice. 7 It should be pointed out that the design of the measurement matrix is beyond the scope of this paper; we mainly focus on the reconstruction of hyperspectral CS. At present, many popular reconstruction algorithms have been proposed.…”

Section: Introductionmentioning

confidence: 99%

Exploring error of linear mixed model for hyperspectral image reconstruction from spectral compressive sensing

Wang

et al. 2019

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

Exploring error of linear mixed model for hyperspectral image reconstruction from spectral compressive sensing," J.Abstract. Linear mixed model (LMM) has been extensively applied for hyperspectral compressive sensing (CS) in recent years. However, the error introduced by LMM that limits the reconstruction performance has not been given full consideration. We propose an algorithm for hyperspectral CS based on LMM under the assumption of known endmembers. At the sampling stage, only spectral compressive sampling is carried out to keep the abundance information as much as possible. At the reconstruction stage, the proposed algorithm estimates abundance by using linear unmixing from the spectral observed data. Moreover, the model error introduced by LMM is explored; a joint convex optimization scheme for estimation of both abundance and model error is established and solved by the alternating iteration approach to achieve the optimal reconstruction. Experimental results on a real hyperspectral dataset demonstrate that the proposed algorithm significantly outperforms the other state-of-the-art hyperspectral CS algorithms.

show abstract

“…This property is commonly utilized to restore images with preserved edges and main features when only incomplete, noisy, or blurred versions of them are available [40]- [42]. It is also known that minimizing the total-variation of an image usually leads to a better recovery performance compared with minimizing the ℓ 1 -norm of the wavelet coefficients of the image [14], [43]. Based on this knowledge, we recover the hyperspectral data from the incomplete and noisy observations by solving the following convex minimization problem:…”

Section: Proposed Algorithmmentioning

confidence: 99%

“…A mathematical tool popularly utilized for the purpose is the theory of compressive sensing [10]- [12], which relies on the assumption that [13], to reduce the sensing complexity and capture time for hyperspectral imaging while maintaining an acceptable reconstruction performance. Among them are the works of [14]- [24].…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Image Recovery via Hybrid Regularization

Arablouei

Hoog

2016

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Natural images tend to mostly consist of smooth regions with individual pixels having highly correlated spectra. This information can be exploited to recover hyperspectral images of natural scenes from their incomplete and noisy measurements. To perform the recovery while taking full advantage of the prior knowledge, we formulate a composite cost function containing a square-error data-fitting term and two distinct regularization terms pertaining to spatial and spectral domains. The regularization for the spatial domain is the sum of total-variation of the image frames corresponding to all spectral bands. The regularization for the spectral domain is the l1-norm of the coefficient matrix obtained by applying a suitable sparsifying transform to the spectra of the pixels. We use an accelerated proximal-subgradient method to minimize the formulated cost function. We analyze the performance of the proposed algorithm and prove its convergence. Numerical simulations using real hyperspectral images exhibit that the proposed algorithm offers an excellent recovery performance with a number of measurements that is only a small fraction of the hyperspectral image data size. Simulation results also show that the proposed algorithm significantly outperforms an accelerated proximal-gradient algorithm that solves the classical basis-pursuit denoising problem to recover the hyperspectral image

show abstract

Compressive sensing and adaptive direct sampling in hyperspectral imaging

Cited by 57 publications

References 25 publications

Multispectral Transforms Using Convolution Neural Networks for Remote Sensing Multispectral Image Compression

Multispectral Transforms Using Convolution Neural Networks for Remote Sensing Multispectral Image Compression

Exploring error of linear mixed model for hyperspectral image reconstruction from spectral compressive sensing

Hyperspectral Image Recovery via Hybrid Regularization

Contact Info

Product

Resources

About