Convolutional Autoencoder for Blind Hyperspectral Image Unmixing

Ranasinghe, Yasiru; Herath, S.; Weerasooriya, H.M.H.K.; Ekanayake, M. P. B.; Godaliyadda, G. M. R. I.; Ekanayake, Parakrama; Herath, Vijitha

doi:10.1109/iciis51140.2020.9342727

Cited by 18 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…New strategies based on supervised and unsupervised machine learning and optimization techniques have been developed to improve the endmember extraction algorithm's accuracy (Xu et al, 2019;Shah et al, 2020). The majority of the unsupervised approaches are based on autoencoders for endmember extraction and estimation of the fractional abundance maps (Palsson et al, 2020;Ranasinghe et al, 2020;Hadi et al, 2022). The autoencoders used to address the SU framework are configured as nonsymmetrical models, where the encoder has more degrees of freedom in the design to add more layers.…”

Section: Figurementioning

confidence: 99%

“…To illustrate the differences in the architectural design of models for SU, some architectures are presented as follows. Ranasinghe et al (2020) proposed a convolutional autoencoder, where the encoder performs three convolutional operations, flatten and dense operations; the last dense layer is set to equal the number of endmembers. The encoder extracts the endmembers, and the last layer performs the reconstruction for the abundance map proposed a deep learning model with three stages.…”

Section: Figurementioning

confidence: 99%

“…Notably, the first architecture integrates ASC constraints using augmented matrices, and the shallow autoencoder utilizes the softmax function in the final layer. On the other hand, the utilization of convolutional architectures in Palsson et al (2020); Ranasinghe et al (2020); Gao et al (2022); Rasti et al (2022) enables the exploitation of spatial and spectral feature extraction. These architectures utilize SAD and MSE and perform abundance estimation by employing loss functions.…”

Section: Publication Reviewmentioning

confidence: 99%

See 2 more Smart Citations

A blind convolutional deep autoencoder for spectral unmixing of hyperspectral images over waterbodies

Alfaro-Mejía,

Manian,

Ortiz

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

Harmful algal blooms have dangerous repercussions for biodiversity, the ecosystem, and public health. Automatic identification based on remote sensing hyperspectral image analysis provides a valuable mechanism for extracting the spectral signatures of harmful algal blooms and their respective percentage in a region of interest. This paper proposes a new model called a non-symmetrical autoencoder for spectral unmixing to perform endmember extraction and fractional abundance estimation. The model is assessed in benchmark datasets, such as Jasper Ridge and Samson. Additionally, a case study of the HSI2 image acquired by NASA over Lake Erie in 2017 is conducted for extracting optical water types. The results using the proposed model for the benchmark datasets improve unmixing performance, as indicated by the spectral angle distance compared to five baseline algorithms. Improved results were obtained for various metrics. In the Samson dataset, the proposed model outperformed other methods for water (0.060) and soil (0.025) endmember extraction. Moreover, the proposed method exhibited superior performance in terms of mean spectral angle distance compared to the other five baseline algorithms. The non-symmetrical autoencoder for the spectral unmixing approach achieved better results for abundance map estimation, with a root mean square error of 0.091 for water and 0.187 for soil, compared to the ground truth. For the Jasper Ridge dataset, the non-symmetrical autoencoder for the spectral unmixing model excelled in the tree (0.039) and road (0.068) endmember extraction and also demonstrated improved results for water abundance maps (0.1121). The proposed model can identify the presence of chlorophyll-a in waterbodies. Chlorophyll-a is an essential indicator of the presence of the different concentrations of macrophytes and cyanobacteria. The non-symmetrical autoencoder for spectral unmixing achieves a value of 0.307 for the spectral angle distance metric compared to a reference ground truth spectral signature of chlorophyll-a. The source code for the proposed model, as implemented in this manuscript, can be found at https://github.com/EstefaniaAlfaro/autoencoder_owt_spectral.git.

show abstract

Section: Figurementioning

confidence: 99%

Section: Figurementioning

confidence: 99%

Section: Publication Reviewmentioning

confidence: 99%

See 1 more Smart Citation

A blind convolutional deep autoencoder for spectral unmixing of hyperspectral images over waterbodies

Alfaro-Mejía,

Manian,

Ortiz

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

show abstract

“…Furthermore, accurate HSI classification has been performed via an improved version of the standard non-negative matrix factorization (NMF) algorithm incorporating fundamental notions of independence (Benachir et al, 2013;Sun et al, 2017). Though existing techniques such as NMF-based unmixing (Rajabi and Ghassemian, 2013;Rathnayake et al, 2020;Wang et al, 2016) or autoencoder architecture based unmixing (Hua et al, 2020;Qi et al, 2020;Ranasinghe et al, 2020) are superior at extracting the endmembers and estimating the corresponding abundances, these algorithms require the knowledge about the number of endmembers to extract which was not available in the first place. Besides, for an algorithm to find feasible survey locations for a particular mineral through single-target identification, such information is superfluous.…”

Section: Related Workmentioning

confidence: 99%

Single-target mineral detection with site-specific endmember extraction for survey points identification: A case study of Jaffna, Sri Lanka

Ranasinghe¹,

Weerasooriya²,

Herath³

et al. 2021

Preprint

View full text Add to dashboard Cite

As field surveys used for manual lithological mapping are costly and time-consuming, digital lithological mapping (DLM) that utilizes remotely sensed spectral imaging provides a viable and economical alternative. Generally, DLM has been performed using spectral imaging with the use of laboratory-generated generic endmember signatures. However, the use of generic signatures is error-prone due to the presence of site-specific impurification processes. To that end, this paper proposes generating a single-target abundance mineral map for DLM, where the generated map can further be used as a guide for the selection or avoidance of a field survey. For that, a stochastic cancellation-based methodology was used to generate a site-specific endemic signature for the mineral in concern to reduce the inclusive nature otherwise present in DLM. Here, single-target detection allows the generation of a more accurate site-specific signature for lithological mapping as opposed to multi-target detection. Furthermore, a soil pixel alignment strategy to visualize the relative purity level of the target mineral has been introduced in the proposed work. Then, for the method validation, mapping of limestone deposits in the Jaffna peninsula of Sri Lanka was conducted as the case study using satellite-based spectral imaging as the input. It was observed that despite the low signal-to-noise ratio of the input hyperspectral data the proposed methodology was able to robustly extract the rich information contained in the input data. Further, a field survey was conducted to collect soil samples of four sites chosen by the proposed DLM from the Jaffna peninsula as an algorithm validation and to demonstrate the application of the proposed solution. The proposed abundance threshold of 0.1 -a hard decision boundary for limestone presence and absence -coincided with the industrial standard X-ray diffraction (XRD) threshold of 5 % for the mineral presence. The results of the XRD test validated the use of the algorithm in the selection of sites to be surveyed, hence could avoid conducting a costly field survey on the assumption of the existence of a mineral. Then a more rigorous survey may be performed, if required, only if the proposed digital lithological survey affirms.

show abstract

“…• The decoder should be limited to a single layer network with linear activations to ensure the resemblance to matrix decomposition [57], [58].…”

Section: Introductionmentioning

confidence: 99%

GAUSS: Guided Encoder-Decoder Architecture for Hyperspectral Unmixing with Spatial Smoothness

Ranasinghe¹,

Weerasooriya²,

Godaliyadda³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In recent hyperspectral unmixing (HU) literature, the application of deep learning (DL) has become more prominent, especially with the autoencoder (AE) architecture. We propose a split architecture and use a pseudo-ground truth for abundances to guide the 'unmixing network' (UN) optimization. Preceding the UN, an 'approximation network' (AN) is proposed, which will improve the association between the centre pixel and its neighbourhood. Hence, it will accentuate spatial correlation in the abundances as its output is the input to the UN and the reference for the 'mixing network' (MN). In the Guided Encoder-Decoder Architecture for Hyperspectral Unmixing with Spatial Smoothness (GAUSS), we proposed using one-hot encoded abundances as the pseudo-ground truth to guide the UN; computed using the k-means algorithm to exclude the use of prior HU methods. Furthermore, we release the single-layer constraint on MN by introducing the UN generated abundances in contrast to the standard AE for HU. Secondly, we experimented with two modifications on the pre-trained network using the GAUSS method. In GAUSS blind , we have concatenated the UN and the MN to back-propagate the reconstruction error gradients to the encoder. Then, in the GAUSSprime, abundance results of a signal processing (SP) method with reliable abundance results were used as the pseudo-ground truth with the GAUSSarchitecture. According to quantitative and graphical results for four experimental datasets, the three architectures either transcended or equated the performance of existing HU algorithms from both DL and SP domains.

show abstract

Convolutional Autoencoder for Blind Hyperspectral Image Unmixing

Cited by 18 publications

References 17 publications

A blind convolutional deep autoencoder for spectral unmixing of hyperspectral images over waterbodies

A blind convolutional deep autoencoder for spectral unmixing of hyperspectral images over waterbodies

Single-target mineral detection with site-specific endmember extraction for survey points identification: A case study of Jaffna, Sri Lanka

GAUSS: Guided Encoder-Decoder Architecture for Hyperspectral Unmixing with Spatial Smoothness

Contact Info

Product

Resources

About