Linear Spectral Unmixing Research Articles

Monitoring the coastal zone using earth observation: application of linear spectral unmixing to coastal dune systems in Wales

AbstractCoastal sand dune systems across temperate Europe are presently characterized by a high level of ecological stabilization and a subsequent loss of biological diversity. The use of continuous monitoring within these systems is vital to the preservation of species richness, particularly with regard to the persistence of early stage pioneer species dependent on a strong sediment supply. Linear spectral unmixing was applied to archived Landsat data (1975–2014) and historical aerial photography (1941–1962) for monitoring bare sand (BS) cover dynamics as a proxy for ecological dune stabilization. Using this approach, a time series of change was calculated for Kenfig Burrows, a 6‐km2 stabilized dune system in South Wales, during 1941–2014. The time series indicated that a rapid level of stabilization had occurred within the study area over a period of 75 years. Accuracy assessment of the data indicated the suitability of medium‐resolution imagery with an RMSE of <10% across all images and a difference of <3% between observed and predicted BS area. Temporal resolution was found to be a significant factor in the representation of BS cover with fluctuations occurring on a sub‐decadal scale, outside of the margin of error introduced through the use of medium‐resolution Landsat imagery. This study demonstrates a tractable approach for mapping and monitoring ecologically sensitive regions at a sub‐Landsat pixel level.

Linear spectral unmixing using endmember coexistence rules and spatial correlation

ABSTRACTMixed pixels are often formed when surface materials are smaller than the spatial resolution of a sensor, or two or more ground features fall within a pixel. Spectral unmixing, decomposing a mixed pixel into a set of endmembers and their corresponding abundance fractions, is an important method for extracting the underlying spectral and spatial information from remote sensing images. Recent studies have shown that it is difficult to increase the accuracy of unmixing using single pixel processing. Here, we suggest combining information on the fundamental interrelations of ground components and a priori knowledge on how ground components co-exist or exclude each other according to general geographic and geomorphic relations with spectral information may allow improved unmixing. Therefore, we propose a novel spectral unmixing method to estimate endmember abundances based on linear spectral mixing model with endmember coexistence rules and spatial correlation (LSMM-R&C). This method was implemented by incorporating endmember coexistence rules along with spatial correlation into a weighted least square method. Experiments with both synthetic and real satellite images were carried out to verify the proposed method, and its performance was also evaluated in comparison to the commonly used LSMM (linear spectral mixture method), LAU (local adaptive unmixing), ISU (iterative spectral unmixing) and ISMA (iterative spectral mixture analysis) methods. LSMM-R&C showed the smallest error, and was more effective at revealing the detailed spatial distribution of endmembers’ abundance, showing high potential for solving the problem of spatial heterogeneity among neighbouring pixels.

Demystifying autofluorescence with excitation-scanning hyperspectral imaging.

Autofluorescence has historically been considered a nuisance in medical imaging. Many endogenous fluorophores, specifically, collagen, elastin, NADH, and FAD, are found throughout the human body. Diagnostically, these signals can be prohibitive since they can outcompete signals introduced for diagnostic purposes. Recent advances in hyperspectral imaging have allowed the acquisition of significantly more data in a shorter time period by scanning the excitation spectra of fluorophores. The reduced acquisition time and increased signal-to-noise ratio allow for separation of significantly more fluorophores than previously possible. Here, we propose to utilize excitation-scanning of autofluorescence to examine tissues and diagnose pathologies. Spectra of autofluorescent molecules were obtained using a custom inverted microscope (TE-2000, Nikon Instruments) with a Xe arc lamp and thin film tunable filter array (VersaChrome, Semrock, Inc.) Scans utilized excitation wavelengths from 360 nm to 550 nm in 5 nm increments. The resultant spectra were used to examine hyperspectral image stacks from various collaborative studies, including an atherosclerotic rat model and a colon cancer study. Hyperspectral images were analyzed with ENVI and custom Matlab scripts including linear spectral unmixing (LSU) and principal component analysis (PCA). Initial results suggest the ability to separate the signals of endogenous fluorophores and measure the relative concentrations of fluorophores among healthy and diseased states of similar tissues. These results suggest pathology-specific changes to endogenous fluorophores can be detected using excitation-scanning hyperspectral imaging. Future work will expand the library of pure molecules and will examine more defined disease states.

Nonlinear Hyperspectral Unmixing Based on Geometric Characteristics of Bilinear Mixture Models

Recently, many nonlinear spectral unmixing algorithms that use various bilinear mixture models (BMMs) have been proposed. However, the high computational complexity and intrinsic collinearity between true endmembers and virtual endmembers considerably decrease these algorithms’ unmixing performances. In this paper, we come up with a novel abundance estimation algorithm based on the BMMs. Motivated by BMMs’ geometric characteristics that are related to collinearity, we conduct a unique nonlinear vertex <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${p}$ </tex-math></inline-formula> to replace all the virtual endmembers. Unlike the virtual endmembers, this vertex <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${p}$ </tex-math></inline-formula> actually works as an additional true endmember that gives affine representations of pixels with other true endmembers. When the pixels’ normalized barycentric coordinates with respect to true endmembers are obtained, they will be directly projected to be their approximate linear mixture components, which removes the collinearity effectively and enables further linear spectral unmixing. After that, based on the analysis of projection bias, two strategies using the projected gradient algorithm and a traditional linear spectral unmixing algorithm, respectively, are provided to correct the bias and estimate more accurate abundances. The experimental results on simulated and real hyperspectral data show that the proposed algorithm performs better compared with both traditional and state-of-the-art spectral unmixing algorithms. Both the unmixing accuracy and speed have been improved.

Identification of hydrothermal alteration minerals associated with geothermal system using ASTER and Hyperion satellite data: a case study from Yankari Park, NE Nigeria

Concealed and fossilized geothermal systems are not characterized by obvious surface manifestations like hotsprings and fumaroles, therefore, could not be easily identifiable using conventional techniques. In this investigation, the applicability of Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and Hyperion data-sets were evaluated in discriminating hydrothermal alteration minerals associated with geothermal systems as a proxy in identifying subtle Geothermal systems at Yankari Park, Nigeria. Feature-oriented principal component selection, spectral angle mapper, linear spectral unmixing were applied to ASTER data based on spectral characteristics of hydrothermal alteration key minerals for a systematic selective extraction of the information of interest. Analytical imaging and geophysics-developed processing methods were applied to Hyperion data for mapping iron oxide/hydroxide minerals and clay mineral assemblages in hydrothermal alteration zones. The results indicate that ASTER and Hyperion could be complemented for reconnaissance stage of targeting subtle alteration mineral assemblages associated with geothermal systems.

Mapping of Aluminous Rich Laterite Depositions through Hyper Spectral Remote Sensing

Increased dimensionality of the satellite data proves to be very useful for discriminating features with very close spectral matching. Present study concentrates on the retrieval of reflectance spectra from the level one radiometrically corrected data in Koraput district (Orissa) for the Bauxite ore. In the present study, atmospheric correction model FLAASH has been used to retrieve reflectance from the radiance data. Preprocessing of the dataset has been done before applying atmospheric correction on the dataset. Spectral subsetting of noise prone bands has been successfully done. Local destriping of the affected bands has been done using a 3*3 local mean filter. Spectral signatures of samples were derived from the processed data. Spectral signature of each sample and derived features vectors were correlated with the satellite image of the area and distribution of each feature was demarcated. Spatial abundance of each feature was used in preparation of mineral abundance map. Accuracy of the map was assessed using training sets of representative geological units. The mineral abundance mapping using the spectral analysis of the reflectance image involves the endmember collection using the N-Dimensional visualizer tool in ENVI software. Laterite, Bauxite, Iron and silica rich Aluminous laterite soil, Alluvium and Forest were selected as the end members after understanding the geology and analysis of the reflectance image. Various mapping techniques were applied to generate the final classified mineral abundance Map, Linear Spectral Unmixing, Mixture Tune Matched Filtering, Spectral Feature Fitting, Spectral Angle Mapper were the techniques used. Results have revealed the ability of Hyper spectral Remote sensing data for the identification and mapping of Hydrothermal altered products like Bauxite, Aluminous Laterite. This technology can be utilized for targeting minerals in the altered zone.

Methods for sandy land detection based on multispectral remote sensing data

Sandification is becoming a serious threat to the sustainability of human habitation. The potential of remote sensing in sandy land detection has been previously demonstrated, but transitional sandy land is difficult to detect because of vegetation cover. The aim of this study, therefore, was to propose methods for sandy land detection based on mixed pixel decomposition and soil particle composition to determine the effects of vegetation coverage and transitional sandy land, using Zhenglan Banner of China as the study area and GF-1 multispectral images as the main data. Results showed that the pixel purity index (PPI) is a viable indicator for pure endmember extraction for sandy land detection via remote sensing. A linear spectral unmixing (LSU) model was established to distinguish sandy land coverage from vegetation, alkaline land, etc. Results showed that without considering the vegetation proportion, when the endmember proportion of sandy land accounted for >50% of the total (except for the vegetation), a pixel would be detected as sandy land, its extraction accuracy was verified to be 86.42% by field data. The results derived from soil particle composition showed that silt was the best indicator for sandy land detection, and clay was secondary to it. Through Partial Least Squares Analysis (PLSA), the percentage of silt content was determined as the dependent variable; R676 and R774 were selected as independent variables to establish the inversion model according to the model effect weights and VIP value. The minimum Prediction Residual Error Sum of Squares (PRESS) was 0.824 tested by leave one out cross validation. The threshold of silt content was determined finally as 3.5%, namely, when the silt content was <3.5%, the pixel would be classified as sandy land, its extraction accuracy was 80.86% that was verified using the same field data.

Mapping Mangrove Species Using Hyperspectral Data: A Case Study of Pichavaram Mangrove Ecosystem, Tamil Nadu

There are many studies on mangrove mapping, zonal demarcation, landuse and land cover changes, and also topics like its loss and restoration using multispectral data. In the recent past, advanced remote-sensing data have been used for species discrimination, mapping, etc. The current research aims at identifying and mapping mangroves species along Pichavarm coast of Tamil Nadu, India, using hyperspectral remote-sensing data. The study attempts to map the species by generating the reference spectra from the existing reports and research papers, as surrogate to expensive field work in conjunction with Hyperion data of January 2013. Image was pre-processed followed by Minimum Noise Fraction (MNF) and Pixel Purity Index (PPI). The output of MNF and PPI has been analysed by visualizing it in n-dimensions for end-member extraction. There were eleven spectra taken from the end-members, which were matched with reference spectra. The spectra—matched—, have been used as an input for classification of data with classifiers like Spectral Angle Mapper (SAM), Spectral Feature Fitting (SFF) and Spectral Information Diversion (SID) to identify and map mangroves species. Further to monitor the exact presence of the species at sub-pixel level, linear spectral un-mixing (LSU) was also performed. The study found SAM with LSU as the best approach for mangrove species mapping in Pichavaram coast.

Mineralogy identification through linear unmixing of blast hole geochemistry

ABSTRACTValidating and updating geological models is important in mining to reduce the errors between the predicted and mined values. In the Banded Iron Formation-hosted iron ore deposits in the Hamersley Ranges Western Australia, geological models are based on widely spaced exploration data, resulting in errors on smaller scales. Closely spaced chemical assays from production blast hole drilling become available for updating models, but do not routinely provide mineralogy. This study identifies mineralogy by applying linear unmixing methods to chemical assays. Linear spectral mixture analysis (LSMA), extended linear mixing model (ELMM) and spectral unmixing within a multi-task GP framework (SUGP) successfully identified the mineralogy in synthetic mixtures, with RMSEs of 13.75%, 11.78% and 11.97%. LSMA had consistently poorer results. SUGP was more accurate when using fewer endmember examples. ELMM was easier to train and required less processing power. Therefore either SUGP or ELMM should be chosen to identify the mineralogy depending on the application.

Hyperspectral and Multispectral Image Fusion Based on Local Low Rank and Coupled Spectral Unmixing

Hyperspectral images (HSIs) usually have high spectral and low spatial resolution. Conversely, multispectral images (MSIs) usually have low spectral and high spatial resolution. The fusion of HSI and MSI aims to create spectral images with high spectral and spatial resolution. In this paper, we propose a fusion algorithm by combining linear spectral unmixing with the local low-rank property. By taking advantage of the local low-rank property, we first partition the corresponding spectral image into patches. For each patch pair, we cast the fusion problem as a coupled spectral unmixing problem that extracts the abundance and the endmembers of MSI and HSI, respectively. It then updates the abundance and the endmember through an alternating update algorithm. In fact, the convergence of the alternative update algorithm can be mathematically and empirically supported. We also propose a multiscale postprocessing procedure to combine fusion results obtained under different patch sizes. In experiments on three data sets, the proposed fusion algorithms outperformed state-of-the-art fusion algorithms in both spatial and spectral domains.

Reducing the Effect of the Endmembers’ Spectral Variability by Selecting the Optimal Spectral Bands

Variable environmental conditions cause different spectral responses of scene endmembers. Ignoring these variations affects the accuracy of fractional abundances obtained from linear spectral unmixing. On the other hand, the correlation between the bands of hyperspectral data is not considered by conventional methods developed for dealing with spectral variability. In this paper, a novel approach is proposed to simultaneously mitigate spectral variability and reduce correlation among different endmembers in hyperspectral datasets. The idea of the proposed method is to utilize the angular discrepancy of bands in the Prototype Space (PS), which is constructed using the endmembers of the image. Using the concepts of PS, in which each band is treated as a space point, we proposed a method to identify independent bands according to their angles. The proposed method comprised two main steps. In the first step, which aims to alleviate the spectral variability issue, image bands are prioritized based on their standard deviations computed over some sets of endmembers. Independent bands are then recognized in the prototype space, employing the angles between the prioritized bands. Finally, the unmixing process is done using the selected bands. In addition, the paper presents a technique to form a spectral library of endmembers’ variability (sets of endmembers). The proposed method extracts endmembers sets directly from the image data via a modified version of unsupervised spatial–spectral preprocessing. The performance of the proposed method was evaluated by five simulated images and three real hyperspectral datasets. The experiments show that the proposed method—using both groups of spectral variability reduction methods and independent band selection methods—produces better results compared to the conventional methods of each group. The improvement in the performance of the proposed method is observed in terms of more appropriate bands being selected and more accurate fractional abundance values being estimated.

Mapping man-induced soil degradation in Armenia's high mountain pastures through remote sensing methods: A case study

The research goal was measuring the precision of linear spectral unmixing (LSU), and NDVI-SMA methods applied to QuickBird imagery, estimating soil surface components related to degradation (fractional vegetation cover-FVC, bare soils fractions, surface rock cover) and determining the appropriateness of these methods for mapping man-induced soil degradation in Armenia's highland pastures. The accuracy assessment was done by comparing between RS derived land cover abundance and the ground truth data. Significant regression was established between ground truth FVC and both NDVI-LSU and LSU - produced vegetation abundance data (R2 = 0.636, R2 = 0.625, respectively). For bare soil fractions, linear regression produced a general coefficient of determination R2 = 0.708. Because of poor spectral resolution of QuickBird imagery LSU failed with the surface rock abundance (R2 = 0.015) assessment. The outcomes show that the proposed method of man-induced soil degradation assessment through FVC, bare soil fractions, and field data adequately reflects the current status of soil degradation throughout the studied pasture site.

Hybrid Spectral Unmixing: Using Artificial Neural Networks for Linear/Non-Linear Switching

Spectral unmixing is a key process in identifying spectral signature of materials and quantifying their spatial distribution over an image. The linear model is expected to provide acceptable results when two assumptions are satisfied: (1) The mixing process should occur at macroscopic level and (2) Photons must interact with single material before reaching the sensor. However, these assumptions do not always hold and more complex nonlinear models are required. This study proposes a new hybrid method for switching between linear and nonlinear spectral unmixing of hyperspectral data based on artificial neural networks. The neural networks was trained with parameters within a window of the pixel under consideration. These parameters are computed to represent the diversity of the neighboring pixels and are based on the Spectral Angular Distance, Covariance and a non linearity parameter. The endmembers were extracted using Vertex Component Analysis while the abundances were estimated using the method identified by the neural networks (Vertex Component Analysis, Fully Constraint Least Square Method, Polynomial Post Nonlinear Mixing Model or Generalized Bilinear Model). Results show that the hybrid method performs better than each of the individual techniques with high overall accuracy, while the abundance estimation error is significantly lower than that obtained using the individual methods. Experiments on both synthetic dataset and real hyperspectral images demonstrated that the proposed hybrid switch method is efficient for solving spectral unmixing of hyperspectral images as compared to individual algorithms.

Sub-pixel vs. super-pixel-based greenspace mapping along the urban–rural gradient using high spatial resolution Gaofen-2 satellite imagery: a case study of Haidian District, Beijing, China

ABSTRACTGreenspace in urban areas is closely related to urban ecosystems, economy, culture, and society. Recently, rapid urban development and expansion are always dominated by a series of human–environment interactions, which can lead to various spatial patterns of urban greenspace especially along the urban–rural gradient. Urban–rural greenspace mapping is therefore of great importance to provide a comprehensive insight for urban planners and managers. In our study, we adopted both the sub-pixel and super-pixel strategies to map the greenspace in Haidian District, Beijing, China. Specifically, the fully constrained linear spectral unmixing and object-based classification methods were implemented as the representatives of sub-pixel and super-pixel strategies, respectively. The high spatial resolution Gaofen-2 multispectral imagery collected in September, 2015 was used in this study. The results showed that the overall accuracies of greenspace mapping by the super-pixel method were higher than those by the sub-pixel method in the selected dense urban, sub-urban, and rural subsets. Obviously, the super-pixel method was more advantageous for mapping greenspace from the high spatial resolution imagery, especially for patches of greenspace in rural and mountain areas. When further comparing these two methods using the medium spatial resolution Landsat-8 imagery, we concluded that the sub-pixel method failed to keep the same levels of greenspace mapping accuracies as those using the high spatial resolution Gaofen-2 imagery but outperformed the super-pixel method especially in the dense urban and sub-urban subsets due to their high degrees of greenspace fragmentation. Furthermore, the sub-pixel method also demonstrated its merits in terms of automation and operability compared to the super-pixel method.

Fragmentation increases wind disturbance impacts on forest structure and carbon stocks in a western Amazonian landscape

Tropical second-growth forests could help mitigate climate change, but the degree to which their carbon potential is achieved will depend on exposure to disturbance. Wind disturbance is common in tropical forests, shaping structure, composition, and function, and influencing successional trajectories. However, little is known about the impacts of extreme winds on second-growth forests in fragmented landscapes, though these ecosystems are often located in mosaics of forest, pasture, cropland, and other land cover types. Indirect evidence suggests that fragmentation increases risk of wind damage in tropical forests, but no studies have found such impacts following severe storms. In this study, we ask whether fragmentation and forest type (old vs. second growth) were associated with variation in wind damage after a severe convective storm in a fragmented production landscape in western Amazonia. We applied linear spectral unmixing to Landsat 8 imagery from before and after the storm, and combined it with field observations of damage to map wind effects on forest structure and biomass. We also used Landsat 8 imagery to map land cover with the goals of identifying old- and second-growth forest and characterizing fragmentation. We used these data to assess variation in wind disturbance across 95,596ha of forest, distributed over 6,110 patches. We find that fragmentation is significantly associated with wind damage, with damage severity higher at forest edges and in edgier, more isolated patches. Damage was also more severe in old-growth than in second-growth forests, but this effect was weaker than that of fragmentation. These results illustrate the importance of considering landscape context in planning tropical forest restoration and natural regeneration projects. Assessments of long-term carbon sequestration potential need to consider spatial variation in disturbance exposure. Where risk of extreme winds is high, minimizing fragmentation and isolation could increase carbon sequestration potential.

Semi-automated scoring of triple-probe FISH in human sperm using confocal microscopy.

Structural and numerical sperm chromosomal aberrations result from abnormal meiosis and are directly linked to infertility. Any live births that arise from aneuploid conceptuses can result in syndromes such as Kleinfelter, Turners, XYY and Edwards. Multi-probe fluorescence in situ hybridization (FISH) is commonly used to study sperm aneuploidy, however manual FISH scoring in sperm samples is labor-intensive and introduces errors. Automated scoring methods are continuously evolving. One challenging aspect for optimizing automated sperm FISH scoring has been the overlap in excitation and emission of the fluorescent probes used to enumerate the chromosomes of interest. Our objective was to demonstrate the feasibility of combining confocal microscopy and spectral imaging with high-throughput methods for accurately measuring sperm aneuploidy. Our approach used confocal microscopy to analyze numerical chromosomal abnormalities in human sperm using enhanced slide preparation and rigorous semi-automated scoring methods. FISH for chromosomes X, Y, and 18 was conducted to determine sex chromosome disomy in sperm nuclei. Application of online spectral linear unmixing was used for effective separation of four fluorochromes while decreasing data acquisition time. Semi-automated image processing, segmentation, classification, and scoring were performed on 10 slides using custom image processing and analysis software and results were compared with manual methods. No significant differences in disomy frequencies were seen between the semi automated and manual methods. Samples treated with pepsin were observed to have reduced background autofluorescence and more uniform distribution of cells. These results demonstrate that semi-automated methods using spectral imaging on a confocal platform are a feasible approach for analyzing numerical chromosomal aberrations in sperm, and are comparable to manual methods. © 2017 International Society for Advancement of Cytometry.

Spectral mineral mapping for characterization of subtle geothermal prospects using ASTER data

In this study, the performance of ASTER data is evaluated for mapping subtle geothermal prospects in an unexplored tropical region having a number of thermal springs. The study employed a simple Decorrelation stretch with specific absorptions to highlight possible alteration zones of interest related to Geothermal (GT) systems. Hydrothermal alteration minerals are subsequently mapped using Spectral Angle Mapper (SAM) and Linear Spectral Unmixing (LSU) algorithms to target representative minerals such as clays, carbonates and AL-OH minerals as indicators of GT activity. The results were validated through field GPS survey, rock sampling and laboratory analysis using latest smart lab X-Ray Diffractometer technology. The study indicates that ASTER broadband satellite data could be used to map subtle GT prospects with the aid of an in-situ verification. However, it also shows that ASTER could not discriminate within specie minerals especially for clays using SWIR bands. Subsequent studies are aimed at looking at both ASTER and Hyperion hyperspectral data in the same area as this could have significant implications for GT resource detection in unmapped aseismic and inaccessible tropical regions using available spaceborne data.

Quantifying Sub-Pixel Surface Water Coverage in Urban Environments Using Low-Albedo Fraction from Landsat Imagery

The problem of mixed pixels negatively affects the delineation of accurate surface water in Landsat Imagery. Linear spectral unmixing has been demonstrated to be a powerful technique for extracting surface materials at a sub-pixel scale. Therefore, in this paper, we propose an innovative low albedo fraction (LAF) method based on the idea of unconstrained linear spectral unmixing. The LAF stands on the “High Albedo-Low Albedo-Vegetation” model of spectral unmixing analysis in urban environments, and investigates the urban surface water extraction problem with the low albedo fraction map. Three experiments are carefully designed using Landsat TM/ETM+ images on the three metropolises of Wuhan, Shanghai, and Guangzhou in China, and per-pixel and sub-pixel accuracies are estimated. The results are compared against extraction accuracies from three popular water extraction methods including the normalized difference water index (NDWI), modified normalized difference water index (MNDWI), and automated water extraction index (AWEI). Experimental results show that LAF achieves a better accuracy when extracting urban surface water than both MNDWI and AWEI do, especially in boundary mixed pixels. Moreover, the LAF has the smallest threshold variations among the three methods, and the fraction threshold of 1 is a proper choice for LAF to obtain good extraction results. Therefore, the LAF is a promising approach for extracting urban surface water coverage.

Using Linear Spectral Unmixing for Subpixel Mapping of Hyperspectral Imagery: A Quantitative Assessment

Subpixel mapping techniques have been widely utilized to determine the spatial distribution of the different land-cover classes in mixed pixels at a subpixel scale by converting low-resolution fractional abundance maps (estimated by a linear mixture model) into a finer classification map. Over the past decades, many subpixel mapping algorithms have been proposed to tackle this problem. It has been obvious that the utilized abundance map has a strong impact on the subsequent subpixel mapping procedure. However, limited attention has been given to the impact of the different aspects in the spectral unmixing model on the subpixel mapping performance. In this paper, a detailed quantitative assessment of different aspects in linear spectral mixture analysis, such as the criteria used to determine the types of pixels, the abundance sum-to-one constraint in the unmixing, and the accuracy of the utilized abundance maps, is investigated. This is accomplished by designing an experimental procedure with replaceable components. A total of six hyperspectral images (four synthetic and two real) were utilized in our experiments. By investigating these critical issues, we can further improve the performance of subpixel mapping techniques.

A GPU-Based Processing Chain for Linearly Unmixing Hyperspectral Images

Linear spectral unmixing is one of the nowadays hottest research topics within the hyperspectral imaging community, being a proof of this fact the vast amount of papers that can be found in the scientific literature about this challenging task. A subset of these works is devoted to the acceleration of previously published unmixing algorithms for application under tight time constraints. For this purpose, hyperspectral unmixing algorithms are typically implemented onto high-performance computing architectures in which the operations involved are executed in parallel, which conducts to a reduction in the time required for unmixing a given hyperspectral image with respect to the sequential version of these algorithms. The speedup factors that can be achieved by means of these high-performance computing platforms heavily depend on the inherent level of parallelism of the algorithms to be executed onto them. However, the majority of the state-of-the-art unmixing algorithms were not originally conceived for being parallelized in an ulterior stage, which clearly restricts the amount of acceleration that can be reached. As far as advanced hyperspectral sensors have increasingly high spatial, spectral, and temporal resolutions, it is hence mandatory to follow a new approach that consists of developing a new class of highly parallel unmixing solutions that can take full advantage of the characteristics of nowadays high-performance computing architectures. This paper represents a step forward toward this direction as it proposes a new parallel algorithm for fully unmixing a hyperspectral image together with its implementation onto two different NVIDIA graphic processing units (GPUs). The results obtained reveal that our proposal is able to unmix hyperspectral images with very different spatial patterns and size better and much faster than the best GPU-based unmixing chains up-to-date published, with independence of the characteristics of the selected GPU.