Algoritmo de Descomposición Ciega Basado en el Modelo de Mezcla Multi-Lineal

2021 ◽  
Author(s):  
◽  
J. N. Mendoza Chavarría
Keyword(s):  
State Of The Art ◽  
Hyperspectral Data ◽  
Spectral Unmixing ◽  
Mixing Models ◽  
Nonlinear Interactions ◽  
Estimation Errors ◽  
Spectral Signatures ◽  
Scattering Effects ◽  
Linear Unmixing ◽  
Multiple Light Scattering

Spectral unmixing has proven to be a great tool for the analysis of hyperspectral data, with linear mixing models (LMMs) being the most used in the literature. Nevertheless, due to the limitations of the LMMs to accurately describe the multiple light scattering effects in multi and hyperspectral imaging, new mixing models have emerged to describe nonlinear interactions. In this paper, we propose a new nonlinear unmixing algorithm based on a multilinear mixture model called Non-linear Extended Blind Endmember and Abundance Extraction (NEBEAE), which is based on a linear unmixing method established in the literature. The results of this study show that proposed method decreases the estimation errors of the spectral signatures and abundance maps, as well as the execution time with respect the state of the art methods.

scholarly journals Parameterized Nonlinear Least Squares for Unsupervised Nonlinear Spectral Unmixing

2019 ◽  
Vol 11 (2) ◽  
pp. 148 ◽  
Author(s):  
Risheng Huang ◽  
Xiaorun Li ◽  
Haiqiang Lu ◽  
Jing Li ◽  
Liaoying Zhao
Keyword(s):  
Least Squares ◽  
Optimization Problem ◽  
Nonnegative Matrix Factorization ◽  
Optimization Problems ◽  
State Of The Art ◽  
Synthetic Data ◽  
Nonnegative Matrix ◽  
Nonlinear Least Squares ◽  
Hyperspectral Data ◽  
Spectral Unmixing

This paper presents a new parameterized nonlinear least squares (PNLS) algorithm for unsupervised nonlinear spectral unmixing (UNSU). The PNLS-based algorithms transform the original optimization problem with respect to the endmembers, abundances, and nonlinearity coefficients estimation into separate alternate parameterized nonlinear least squares problems. Owing to the Sigmoid parameterization, the PNLS-based algorithms are able to thoroughly relax the additional nonnegative constraint and the nonnegative constraint in the original optimization problems, which facilitates finding a solution to the optimization problems . Subsequently, we propose to solve the PNLS problems based on the Gauss–Newton method. Compared to the existing nonnegative matrix factorization (NMF)-based algorithms for UNSU, the well-designed PNLS-based algorithms have faster convergence speed and better unmixing accuracy. To verify the performance of the proposed algorithms, the PNLS-based algorithms and other state-of-the-art algorithms are applied to synthetic data generated by the Fan model and the generalized bilinear model (GBM), as well as real hyperspectral data. The results demonstrate the superiority of the PNLS-based algorithms.

scholarly journals Using a Panchromatic Image to Improve Hyperspectral Unmixing

2020 ◽  
Vol 12 (17) ◽  
pp. 2834
Author(s):  
Simon Rebeyrol ◽  
Yannick Deville ◽  
Véronique Achard ◽  
Xavier Briottet ◽  
Stephane May
Keyword(s):  
Matrix Factorization ◽  
State Of The Art ◽  
Hyperspectral Data ◽  
Spectral Unmixing ◽  
Endmember Extraction ◽  
Local Approach ◽  
Hyperspectral Unmixing ◽  
Panchromatic Image ◽  
Art Methods ◽  
Non Negative Matrix Factorization

Hyperspectral unmixing is a widely studied field of research aiming at estimating the pure material signatures and their abundance fractions from hyperspectral images. Most spectral unmixing methods are based on prior knowledge and assumptions that induce limitations, such as the existence of at least one pure pixel for each material. This work presents a new approach aiming to overcome some of these limitations by introducing a co-registered panchromatic image in the unmixing process. Our method, called Heterogeneity-Based Endmember Extraction coupled with Local Constrained Non-negative Matrix Factorization (HBEE-LCNMF), has several steps: a first set of endmembers is estimated based on a heterogeneity criterion applied on the panchromatic image followed by a spectral clustering. Then, in order to complete this first endmember set, a local approach using a constrained non-negative matrix factorization strategy, is proposed. The performance of our method, in regards of several criteria, is compared to those of state-of-the-art methods obtained on synthetic and satellite data describing urban and periurban scenes, and considering the French HYPXIM/HYPEX2 mission characteristics. The synthetic images are built with real spectral reflectances and do not contain a pure pixel for each endmember. The satellite images are simulated from airborne acquisition with the spatial and spectral features of the mission. Our method demonstrates the benefit of a panchromatic image to reduce some well-known limitations in unmixing hyperspectral data. On synthetic data, our method reduces the spectral angle between the endmembers and the real material spectra by 46% compared to the Vertex Component Analysis (VCA) and N-finder (N-FINDR) methods. On real data, HBEE-LCNMF and other methods yield equivalent performance, but, the proposed method shows more robustness over the data sets compared to the tested state-of-the-art methods. Moreover, HBEE-LCNMF does not require one to know the number of endmembers.

scholarly journals Stochastic Perturbations on Low-Rank Hyperspectral Data for Image Classification

2021 ◽  
Vol 5 (1) ◽  
pp. 1
Author(s):  
Alex Sumarsono ◽  
Farnaz Ganjeizadeh ◽  
Ryan Tomasi
Keyword(s):  
Machine Learning ◽  
Least Squares ◽  
Classification Accuracy ◽  
State Of The Art ◽  
Hyperspectral Data ◽  
Low Rank ◽  
Stochastic Perturbations ◽  
Sparse Decomposition ◽  
Spectral Signatures ◽  
Material Substances

Hyperspectral imagery (HSI) contains hundreds of narrow contiguous bands of spectral signals. These signals, which form spectral signatures, provide a wealth of information that can be used to characterize material substances. In recent years machine learning has been used extensively to classify HSI data. While many excellent HSI classifiers have been proposed and deployed, the focus has been more on the design of the algorithms. This paper presents a novel data preprocessing method (LRSP) to improve classification accuracy by applying stochastic perturbations to the low-rank constituent of the dataset. The proposed architecture is composed of a low-rank and sparse decomposition, a degradation function and a constraint least squares filter. Experimental results confirm that popular state-of-the-art HSI classifiers can produce better classification results if supplied by LRSP-altered datasets rather than the original HSI datasets. 

scholarly journals How Hyperspectral Image Unmixing and Denoising Can Boost Each Other

2020 ◽  
Vol 12 (11) ◽  
pp. 1728
Author(s):  
Behnood Rasti ◽  
Bikram Koirala ◽  
Paul Scheunders ◽  
Pedram Ghamisi
Keyword(s):  
Noise Reduction ◽  
Hyperspectral Image ◽  
Optimization Criterion ◽  
Hyperspectral Data ◽  
Spectral Unmixing ◽  
Low Rank ◽  
Hyperspectral Unmixing ◽  
Selection Strategies ◽  
Linear Unmixing ◽  
The One

Hyperspectral linear unmixing and denoising are highly related hyperspectral image (HSI) analysis tasks. In particular, with the assumption of Gaussian noise, the linear model assumed for the HSI in the case of low-rank denoising is often the same as the one used in HSI unmixing. However, the optimization criterion and the assumptions on the constraints are different. Additionally, noise reduction as a preprocessing step in hyperspectral data analysis is often ignored. The main goal of this paper is to study experimentally the influence of noise on the process of hyperspectral unmixing by: (1) investigating the effect of noise reduction as a preprocessing step on the performance of hyperspectral unmixing; (2) studying the relation between noise and different endmember selection strategies; (3) investigating the performance of HSI unmixing as an HSI denoiser; (4) comparing the denoising performance of spectral unmixing, state-of-the-art HSI denoising techniques, and the combination of both. All experiments are performed on simulated and real datasets.

scholarly journals Robust blind spectral unmixing for fluorescence microscopy using unsupervised learning

2019 ◽  
Author(s):  
Tristan D. McRae ◽  
David Oleksyn ◽  
Jim Miller ◽  
Yu-Rong Gao
Keyword(s):  
Fluorescence Microscopy ◽  
Emission Spectra ◽  
Principal Component ◽  
Spectral Unmixing ◽  
Spectral Signatures ◽  
Two Photon Microscopy ◽  
Two Photon ◽  
False Positive Detection ◽  
Linear Unmixing ◽  
Microscopy Images

AbstractDue to the overlapping emission spectra of fluorophores, fluorescence microscopy images often have bleed-through problems, leading to a false positive detection. This problem is almost unavoidable when the samples are labeled with three or more fluorophores, and the situation is complicated even further when imaged under a multiphoton microscope. Several methods have been developed and commonly used by biologists for fluorescence microscopy spectral unmixing, such as linear unmixing, non-negative matrix factorization, deconvolution, and principal component analysis. However, they either require pre-knowledge of emission spectra or restrict the number of fluorophores to be the same as detection channels, which highly limits the real-world applications of those spectral unmixing methods. In this paper, we developed a robust and flexible spectral unmixing method: Learning Unsupervised Means of Spectra (LUMoS), which uses an unsupervised machine learning clustering method to learn individual fluorophores’ spectral signatures from mixed images, and blindly separate channels without restrictions on the number of fluorophores that can be imaged. This method highly expands the hardware capability of two-photon microscopy to simultaneously image more fluorophores than is possible with instrumentation alone. Experimental and simulated results demonstrated the robustness of LUMoS in multi-channel separations of two-photon microscopy images. We also extended the application of this method to background/autofluorescence removal and colocalization analysis. Lastly, we integrated this tool into ImageJ to offer an easy to use spectral unmixing tool for fluorescence imaging. LUMoS allows us to gain a higher spectral resolution and obtain a cleaner image without the need to upgrade the imaging hardware capabilities.

scholarly journals A Supervised Method for Nonlinear Hyperspectral Unmixing

2019 ◽  
Vol 11 (20) ◽  
pp. 2458 ◽  
Author(s):  
Bikram Koirala ◽  
Mahdi Khodadadzadeh ◽  
Cecilia Contreras ◽  
Zohreh Zahiri ◽  
Richard Gloaguen ◽  
...  
Keyword(s):  
Feedforward Neural Networks ◽  
Ground Truth ◽  
Hyperspectral Data ◽  
Spectral Unmixing ◽  
Model Parameters ◽  
Data Set ◽  
Fractional Abundance ◽  
Highly Nonlinear ◽  
Linear Unmixing ◽  
Nonlinear Mixtures

Due to the complex interaction of light with the Earth’s surface, reflectance spectra can be described as highly nonlinear mixtures of the reflectances of the material constituents occurring in a given resolution cell of hyperspectral data. Our aim is to estimate the fractional abundance maps of the materials from the nonlinear hyperspectral data. The main disadvantage of using nonlinear mixing models is that the model parameters are not properly interpretable in terms of fractional abundances. Moreover, not all spectra of a hyperspectral dataset necessarily follow the same particular mixing model. In this work, we present a supervised method for nonlinear spectral unmixing. The method learns a mapping from a true hyperspectral dataset to corresponding linear spectra, composed of the same fractional abundances. A simple linear unmixing then reveals the fractional abundances. To learn this mapping, ground truth information is required, in the form of actual spectra and corresponding fractional abundances, along with spectra of the pure materials, obtained from a spectral library or available in the dataset. Three methods are presented for learning nonlinear mapping, based on Gaussian processes, kernel ridge regression, and feedforward neural networks. Experimental results conducted on an artificial dataset, a data set obtained by ray tracing, and a drill core hyperspectral dataset shows that this novel methodology is very promising.

scholarly journals Hyperspectral Data Simulation (Sentinel-2 to AVIRIS-NG) for Improved Wildfire Fuel Mapping, Boreal Alaska

2021 ◽  
Vol 13 (9) ◽  
pp. 1693
Author(s):  
Anushree Badola ◽  
Santosh K. Panda ◽  
Dar A. Roberts ◽  
Christine F. Waigl ◽  
Uma S. Bhatt ◽  
...  
Keyword(s):  
Land Cover ◽  
Spectral Resolution ◽  
Hyperspectral Image ◽  
Spectral Response ◽  
Spectral Characteristics ◽  
Hyperspectral Data ◽  
Spectral Unmixing ◽  
High Spectral Resolution ◽  
Fuel Mapping ◽  
Sentinel 2

Alaska has witnessed a significant increase in wildfire events in recent decades that have been linked to drier and warmer summers. Forest fuel maps play a vital role in wildfire management and risk assessment. Freely available multispectral datasets are widely used for land use and land cover mapping, but they have limited utility for fuel mapping due to their coarse spectral resolution. Hyperspectral datasets have a high spectral resolution, ideal for detailed fuel mapping, but they are limited and expensive to acquire. This study simulates hyperspectral data from Sentinel-2 multispectral data using the spectral response function of the Airborne Visible/Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) sensor, and normalized ground spectra of gravel, birch, and spruce. We used the Uniform Pattern Decomposition Method (UPDM) for spectral unmixing, which is a sensor-independent method, where each pixel is expressed as the linear sum of standard reference spectra. The simulated hyperspectral data have spectral characteristics of AVIRIS-NG and the reflectance properties of Sentinel-2 data. We validated the simulated spectra by visually and statistically comparing it with real AVIRIS-NG data. We observed a high correlation between the spectra of tree classes collected from AVIRIS-NG and simulated hyperspectral data. Upon performing species level classification, we achieved a classification accuracy of 89% for the simulated hyperspectral data, which is better than the accuracy of Sentinel-2 data (77.8%). We generated a fuel map from the simulated hyperspectral image using the Random Forest classifier. Our study demonstrated that low-cost and high-quality hyperspectral data can be generated from Sentinel-2 data using UPDM for improved land cover and vegetation mapping in the boreal forest.

scholarly journals An original method for tree species classification using multitemporal multispectral and hyperspectral satellite data

Silva Fennica ◽  
2020 ◽  
Vol 54 (2) ◽  
Author(s):  
Olga Grigorieva ◽  
Olga Brovkina ◽  
Alisher Saidov
Keyword(s):  
Tree Species ◽  
Original Method ◽  
Hyperspectral Data ◽  
Species Classification ◽  
Multispectral Data ◽  
Spectral Signatures ◽  
Tree Species Classification ◽  
Temporal Features ◽  
Study Sites ◽  
Near Future

This study proposes an original method for tree species classification by satellite remote sensing. The method uses multitemporal multispectral (Landsat OLI) and hyperspectral (Resurs-P) data acquired from determined vegetation periods. The method is based on an original database of spectral features taking into account seasonal variations of tree species spectra. Changes in the spectral signatures of forest classes are analyzed and new spectral–temporal features are created for the classification. Study sites are located in the Czech Republic and northwest (NW) Russia. The differences in spectral reflectance between tree species are shown as statistically significant in the sub-seasons of spring, first half of summer, and main autumn for both study sites. Most of the errors are related to the classification of deciduous species and misclassification of birch as pine (NW Russia site), pine as mixture of pine and spruce, and pine as mixture of spruce and beech (Czech site). Forest species are mapped with accuracy as high as 80% (NW Russia site) and 81% (Czech site). The classification using multitemporal multispectral data has a kappa coefficient 1.7 times higher than does that of classification using a single multispectral image and 1.3 times greater than that of the classification using single hyperspectral images. Potentially, classification accuracy can be improved by the method when applying multitemporal satellite hyperspectral data, such as in using new, near-future products EnMap and/or HyspIRI with high revisit time.

scholarly journals Identification and mapping of ancient pigments in a Roman Egyptian funerary portrait by application of reflectance and luminescence imaging spectroscopy

2022 ◽  
Vol 10 (1) ◽  
Author(s):  
Roxanne Radpour ◽  
Glenn A. Gates ◽  
Ioanna Kakoulli ◽  
John K. Delaney
Keyword(s):  
Molecular Spectroscopy ◽  
Imaging Spectroscopy ◽  
Chemical Mapping ◽  
Spectral Signatures ◽  
Egyptian Blue ◽  
Non Invasive ◽  
Analytical Protocol ◽  
Linear Unmixing ◽  
New Applications ◽  
Madder Lake

AbstractImaging spectroscopy (IS) is an important tool in the comprehensive technical analysis required of archaeological paintings. The complexity of pigment mixtures, diverse artistic practices and painting technologies, and the often-fragile and weathered nature of these objects render macroscale, non-invasive chemical mapping an essential component of the analytical protocol. Furthermore, the use of pigments such as Egyptian blue and madder lake, featuring diagnostic photoluminescence emission, provides motivation to perform photoluminescence mapping on the macroscale. This work demonstrates and advances new applications of dual-mode imaging spectroscopy and data analysis approaches for ancient painting. Both reflectance (RIS) and luminescence (LIS) modes were utilized for the study of a Roman Egyptian funerary portrait from second century CE Egypt. The first derivative of the RIS image cube was analyzed and found to significantly improve materials separation, identification, and the extent of mapping. Egyptian blue and madder lake were mapped across a decorated surface using their luminescence spectral signatures in the region of 540–1000 nm as endmembers in LIS analyses. Linear unmixing of the LIS endmembers and subsequent derivative analyses resulted in an improved separation and mapping of the luminescence pigments. RIS and LIS studies, combined with complementary, single-spot collection elemental and molecular spectroscopy, were able to successfully characterize the portrait’s painting materials and binding media used by the ancient artist, providing key insight into their material use, stylistic practices, and technological choices.

scholarly journals HYPERSPECTRAL REMOTE SENSING ROLE IN ENHANCING CROP MAPPING: A COMPARISON BETWEEN DIFFERENT SUPERVISED SEGMENTATION ALGORITHMS

2019 ◽  
Vol 1 (1) ◽  
pp. 25-37
Author(s):  
Mohamad M. Awad
Keyword(s):  
Remote Sensing ◽  
Eastern Mediterranean ◽  
Hyperspectral Data ◽  
Spectral Angle Mapper ◽  
New Approach ◽  
Spectral Signatures ◽  
Agriculture Sector ◽  
Supervised Segmentation ◽  
Eastern Mediterranean Basin ◽  
Crop Mapping

In agriculture sector there is need for cheap, fast, and accurate data and technologies to help decision makers to find solutions for many agricultural problems. Many solutions depend significantly on the accuracy and efficiency of the crop mapping and crop yield estimation processes. High resolution spectral remote sensing can improve substantially crop mapping by reducing similarities between different crop types which has similar ecological conditions. This paper presents a new approach of combining a new tool, hyperspectral images and technologies to enhance crop mapping.  The tool includes spectral signatures database for the major crops in the Eastern Mediterranean Basin and other important metadata and processing functions. To prove the efficiency of the new approach, major crops such as “winter wheat” and “spring potato” are mapped using the spectral signatures database in the new tool, three different supervised algorithms, and CHRIS-Proba hyperspectral satellite images. The evaluation of the results showed that deploying different hyperspectral data and technologies can improve crop mapping. The improvements can be noticed with the increase of the accuracy to more than 86% with the use of the supervised algorithm Spectral Angle Mapper (SAM).

Sign in / Sign up

Export Citation Format

Share Document