Fast Hyperspectral Unmixing Using a Multiscale Sparse Regularization

İnce, Taner; Dobigeon, Nicolas

doi:10.1109/lgrs.2022.3217872

Cited by 6 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed S 2 MSU algorithm 2 is compared to several sparse unmixing from the literature: SUnSAL [12], SUnSAL-TV [14], S 2 WSU [16], DRSU-TV [15], MUA [17], FastUn [18] and SUSRLR-TV [19]. Their performance is assessed by computing the signal reconstruction error (SRE) defined as SRE = 10 log 10…”

Section: Resultsmentioning

confidence: 99%

“…Thanks to its simplicity, MUA has a much lower computational complexity than TV based approaches but it may be shown to smooth edges [17]. Recently, fast sparse unmixing (FastUn) builds on this multiscale approach while simultaneously preserving crisp edge details thanks to a spatial discontinuity strategy [18].…”

Section: Introductionmentioning

confidence: 99%

“…The computational complexity of the method is significantly lower than those of competitive algorithms without sacrificing the unmixing accuracy. Even though the recently proposed FastUn algorithm [18] shares some similarities with the proposed S 2 MSU method, it is worth noting that the former does not exploit the spectral redundancy inherent to the hyperspectral data. Contrary to the proposed S 2 MSU method, the design of the FastUn regularization is only motivated by spatially motivated arguments.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Spatial–Spectral Multiscale Sparse Unmixing for Hyperspectral Images

Ince,

Dobigeon

2023

IEEE Geosci. Remote Sensing Lett.

Self Cite

View full text Add to dashboard Cite

We propose a simple yet efficient sparse unmixing method for hyperspectral images. It exploits the spatial and spectral properties of hyperspectral images by designing a new regularization informed by multiscale analysis. The proposed approach consists of two steps. First, a sparse unmixing is conducted on a coarse hyperspectral image resulting from a spatial smoothing of the original data. The estimated coarse abundance map is subsequently used to design two weighting terms summarizing the spatial and spectral properties of the image. They are combined to define a sparse regularization embedded into a unmixing problem associated with the original hyperspectral image at full resolution. The performance of the proposed method is assessed with numerous experiments conducted on synthetic and real data sets. It is shown to compete favorably with state-of-the-art methods from the literature with lower computational complexity.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Spatial–Spectral Multiscale Sparse Unmixing for Hyperspectral Images

Ince,

Dobigeon

2023

IEEE Geosci. Remote Sensing Lett.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The spatial weight factor W is calculated using the coarse abundance map assumed to captures the interpixel spatial structure w i,j = 1 |ǎ i,j | + ϵ where ϵ a small constant introduced to numerical instabilities. This strategy allows not only to promote spatial smoothness in homogeneous regions but also to preserve the details and crisp structures in the final estimated maps [24], contrary to the conventional TV regularization which does not account for edge details.…”

Section: B Estimation Of Abundances At Full Resolutionmentioning

confidence: 99%

A Fast Spatial–Spectral NMF for Hyperspectral Unmixing

İnce

Dobigeon

2023

IEEE Geosci. Remote Sensing Lett.

Self Cite

View full text Add to dashboard Cite

This letter proposes a fast yet efficient method to solve the hyperspectral unmixing problem in the challenging unsupervised context, i.e., when the endmember spectral signatures are unknown. First, a coarse approximation of the hyperspectral image is computed by spatially averaging neighboring pixels, which significantly reduces the amount of pixels to be handled. This reduced set of hyperspectral pixels is unmixed to derive coarse solutions of the unmixing problem, i.e., coarse estimates of the endmember signatures and the corresponding low-resolution abundance maps. Then, the plain resolution abundance maps are estimated from the corresponding hyperspectral image based on the coarse endmember signatures. A sparsity promoting prior exploiting the low resolution map complements the conventional data fitting term to promote spatial smoothness while mitigating the loss of details in the edge areas. Finally, a least square optimization problem is solved to obtain the actual endmember signatures from the hyperspectral image and the abundance maps of plain resolution estimated in the previous step. Numerical experiments show that the proposed method is fast and performs well compared to state-of-the-art approaches from the literature.

show abstract

“…The availability of rich spectral information allows identification of materials. Therefore, it is used in many areas such as mineral detection [2]- [4], target detection [5]- [7], classification of land surfaces [8]- [11] and so on.…”

Section: Introductionmentioning

confidence: 99%

Guided Filter of Random Patches Network and Relaxed-Collaborative-Representation-Based Hyperspectral Image Classification

Dundar,

Ince

2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

Feature extraction and accurate classification are crucial tasks in land-cover classification of hyperspectral image (HSI). We propose guided filter (GF) of random patches network (RPNet) and relaxed collaborative representation (RCR) based HSI classification (HSIC) method called GRR. The shallow and deep features are extracted using RPNet that requires no pretraining stage. In addition to the obtained feature set, the original HSI and extracted features are then filtered by GF to preserve the edge details. After that, all the distinct feature sets are separately concatenated with the original HSI to keep the original structure of the data. The high-dimensional feature sets are then processed by linear discriminant analysis (LDA) to increase class separability and to select the most representative features. Since few train samples are available in HSIC task, efficiency of LDA is improved using superpixel segmentation to generate pseudosamples. As final stage, the reduced-dimension feature sets are classified by the use of superpixel-guided RCR which utilize the resemblance and discrimination of the feature sets efficiently. The extensive experiments on the real HSIs are carried out to validate the efficacy of the proposed method.

show abstract

Fast Hyperspectral Unmixing Using a Multiscale Sparse Regularization

Cited by 6 publications

References 24 publications

Spatial–Spectral Multiscale Sparse Unmixing for Hyperspectral Images

Spatial–Spectral Multiscale Sparse Unmixing for Hyperspectral Images

A Fast Spatial–Spectral NMF for Hyperspectral Unmixing

Guided Filter of Random Patches Network and Relaxed-Collaborative-Representation-Based Hyperspectral Image Classification

Contact Info

Product

Resources

About