A fully constraint abundance estimation algorithm with high accurateness

Spectral unmixing is an important task in remotely sensed hyperspectral data exploitation. The linear mixture model has been widely used to unmix hyperspectral images by identifying a set of pure spectral signatures, called endmembers, and estimating their respective abundances in each pixel of the scene. Several algorithms have been proposed in the recent literature to automatically identify endmembers, even if the original hyperspectral scene does not contain any pure signatures. A popular strategy for endmember identification in highly mixed hyperspectral scenes has been the minimum volume simplex analysis (MVSA), known to be a computationally very expensive algorithm. This algorithm calculates the minimum volume enclosing simplex, as opposed to other algorithms that perform maximum simplex volume analysis (MSVA). The high computational complexity of MVSA, together with its very high memory requirements, has limited its adoption in the hyperspectral imaging community. In this paper, we develop several optimizations to the MVSA algorithm. The main computational task of MVSA is the solution of a quadratic optimization problem with equality and inequality constraints, with the inequality constraints being in the order of the number of pixels multiplied by the number of endmembers. As a result, storing and computing the inequality constraint matrix is highly inefficient. The first optimization presented in this paper uses algebra operations in order to reduce the memory requirements of the algorithm. In the second optimization, we use graphics processing units (GPUs) to effectively solve (in parallel) the quadratic optimization problem involved in the computation of MVSA. In the third optimization, we extend the single GPU implementation to a multi-GPU one, developing a hybrid strategy that distributes the computation while taking advantage of GPU accelerators at each node. The presented optimizations are tested in different analysis scenarios (using both synthetic and real hyperspectral data) and shown to provide state-of-the-art results from the viewpoint of unmixing accuracy and computational performance. The speedup achieved using the full GPU cluster compared to the CPU implementation in tenfold in a real hyperspectral image.

In this paper, we propose a proposed fully constrained least squares (p-FCLS) for spectral unmixing, which assumes that the endmember with the absolute maximum negative abundance by least squares is the disturbing endmember and that can be ignored subsequently, then the other endmembers are returned in the iteration until the abundance of all endmembers are non-negative. To evaluate the effectiveness of the proposed algorithm, hyperspectral data samples by ASD Field spectrometer and the EO-1 Hyperion image are employed. The results illustrate that p-FCLS could reduced time consumption and is more accurate in unmixing hyperspectral mixed spectral and pixel than fully constrained least squares (FCLS) and normalized non-negative matrix factorization (NNMF).

With the advancement in technology, hyperspectral images have potential applications in the field of remote sensing due to their high spectral resolution. Despite the hyperspectral image providing abundant information, its analysis suffers from the problem of high dimensionality. Hence, Dimensionality Reduction (DR) is an essential task in all hyperspectral image analysis. Band Selection, which is one of the DR techniques, is still a challenging issue even though many algorithms have been developed. To provide remedy for this issue, this paper explores a novel approach for band selection using K-means clustering on statistical feature in hyperspectral images. The proposed method of clustering based band selection for DR is simple and accurate. A reliable estimate of number of bands to be selected is provided by Virtual Dimensionality (VD). Informative bands preserving maximum information are selected based on the statistical feature, the variance using K-means Clustering technique. Further, our proposed work involves the utilization of the effectiveness of Simplex Growing Algorithm (SGA) on endmember extraction in association with clustering based band selection. Using Fully Constrained Least Squares (FCLS) method, abundance fraction is estimated based on endmember signatures, which are derived using Endmember Extraction Algorithm (EEA). The proposed work is investigated and compared with that of N-FINDR and Vertex Component Analysis (VCA) algorithms. The performance of the proposed algorithm is evaluated using Root Mean Square Error (RMSE), Spectral Angle Distance (SAD) and computation time. Experimental results show that the proposed clustering based band selection with SGA endmember extraction algorithm reduces the average SAD by 8 to 10 % and the average RMSE by nearly 1 %, compared to that of N-FINDR and VCA algorithms. In terms of computation time, the proposed band selection based DR with SGA algorithm is seven times faster than conventional transform based DR with SGA algorithm.

Hyperspectral imaging is concerned with the measurement, analysis, and interpretation of spectra acquired from a given scene (or specific object) at a short, medium or long distance by an airbone or satellite sensor. Over the last few years, hyperspectral image data sets have been collected for a great amount of locations over the world, using a variety of instruments for Earth observation. Despite the increasing importance of hyperspectral images in remote sensing applications, there is no common repository of hyperspectral data intended to distribute and share hyperspectral data sets in the community. Quite opposite, the hyperspectral data sets which are available for public use are spread among different storage locations and present significant heterogeneity regarding the storage format, associated meta-data (if any), or ground-truth availability. As a result, the development of a standardized hyperspectral data repository is a highly desired goal in the remote sensing community. In this paper, we take a necessary first step towards the development of a digital repository for remotely sensed hyperspectral data. The proposed system allows uploading new hyperspectral data sets along with meta-data, ground-truth and analysis results, with the ultimate goal of sharing publicly available hyperspectral images within the remote sensing community. The database has been designed in order to allow storing relevant information for the hyperspectral data available through the system, including basic image characteristics (width, height, number of bands, format) and more advanced meta-data (ground-truth information, publications in which the data has been used). The current implementation consists of a front-end to ease the management of images through a web interface, thus containing both synthetic and real hyperspectral images from two highly representative instruments, such as NASAs Airborne Visible Infra-Red Imaging Spectrometer (AVIRIS) over the Cuprite Mining District in Nevada. Most importantly, the developed system includes a spectral unmixing-based content based image retrieval (CBIR) functionality which allows searching for images on the spectral unmixing information (spectrally pure components or endmembers and their associated abundances in the scene). This information is stored as meta-data associated to each hyperspectral image instance, and then used to search and retrieve images based on information content. This paper presents the design of the system and a preliminary validation of the unmixing-based retrieval functionality using both synthetic and real hyperspectral images stored in the database.

Signal subspace identification is a crucial step in many hyperspectral processing algorithms such as target detection, change detection, classification, and unmixing. This is due to the fact that a correct dimensionality reduction improves algorithm performances and reduces their complexity and data requirements. This paper introduces a new method for this task which is based on the Ritz values obtained in methods such as the restarted Arnoldi method or the Lanczos method for calculating the eigenvalues and eigenvectors of a given matrix. In particular, it first calculates a high number of eigenvalues and eigenvectors, and a Ritz value per eigenvalue calculated and then, it estimates the dimension of the signal subspace according to the Ritz values obtained. The results obtained with the proposed method for synthetic and real hyperspectral images verify the performance of the introduced methodology to estimate the number of endmembers in different types of hyperspectral images. Moreover, these results are better than the ones obtained, for the same images, with the most popular algorithm for this task in the state of the art, i.e. the HySIME and the Virtual Dimensionality algorithms.

The objective function of classical nonnegative matrix factorization (NMF) is non-convexity, which affects the obtaining of optimal solutions. In this paper, we proposed a NMF algorithm, and this algorithm was based on the constraint of endmember spectral correlation minimization and endmember spectral difference maximization. The size of endmember spectral overall-correlation was measured by the correlation function, and correlation function was defined as the sum of the absolute values of every two correlation coefficient between the spectra. In the difference constraint of the endmember spectra, the mutation of matrix trace was slowed down by introducing the natural logarithm function. Combining the image decomposition error with the influences of endmember spectra, in the objective function the projection gradient was used to achieve NMF. The effectiveness of algorithm was verified by the simulated hyperspectral images and real hyperspectral images.

International audience

Knowing the number of endmembers in a hyperspectral image is a prerequisite for almost all the endmember extraction algorithms and plays a key role for the accuracy of the spectral unmixing. Moreover, in case of data compression, it is important to know the number of endmembers in order to define the appropriate signal subspace. In this paper, a new automated method for estimating the number of endmembers in hyperspectral imagery is proposed, without the need of a priori knowledge. The method is based on the intra-band standard deviation values of the transformed components produced by eigen-based decomposition, and uses a fixed threshold -the same regardless the hyperspectral dataset- to define the optimum signal subspace. The effectiveness of the proposed method is shown using synthetic and real data. Comparison with state-of-the-art methods for the estimation of the number of endmembers is also performed.

Convolutional neural network (CNN) has shown superior classification results for hyperspectral image classification. Due to high dimensionality of the hyperspectral image, usually a feature reduction method is applied to the hyperspectral data before feeding it to the CNN model. Principal component analysis is the most widely used approach to this end. However, it has two main disadvantages. It uses the mean square error measure and ignores the class separability. Moreover, it is a pixel-based method, which does not utilize the spatial information. To deal with these weaknesses, the patch-based nonparametric discriminant analysis method is proposed for hyperspectral image feature reduction in this work. However, due to the complex and heterogeneous nature of the hyperspectral images, the use of fixed square patches is not so appropriate. To deal with this difficulty, a shaping approach is introduced for the generation of adaptive patches with consistent shapes in each local region. The input patches of the CNN model are also weighted according to the spectral similarity and class boundaries information. Moreover, a simple residual CNN model is suggested for the classification of the extracted shaped input patches. The efficiency of the proposed framework is assessed in different cases using various evaluation measures on six real hyperspectral images.

Many remote sensing applications require a high-resolution hyperspectral image. However, resolutions of most hyperspectral imagers are limited to tens of meters. Existing resolution enhancement techniques either acquire additional multispectral band images or use a pan band image. The former poses hardware challenges, whereas the latter has limited performance. In this paper, we present a new resolution enhancement method that only requires a color image. Our approach integrates two newly developed techniques in the area: (1) A hybrid color mapping algorithm, and (2) A Plug-and-Play algorithm for single image super-resolution. Comprehensive experiments using real hyperspectral images are conducted to validate and evaluate the proposed method.

The normal compositional model (NCM) is a well-known and powerful model in hyperspectral unmixing which represents endmembers as independent Gaussian vectors to capture endmember variability. However, the assumption of independent endmembers diminishes the model accuracy because the high degree of correlation between endmembers of a scene and identical sources of variability demonstrate that the endmembers are dependent. This paper proposes a new hyperspectral unmixing algorithm which represents endmembers using dependent Gaussian vectors to estimate abundance fractions. To overcome the higher complexity caused by dependence assumption, this algorithm introduces new independent Gaussian vectors named Base Vectors to represent different endmembers by a weighted linear combination. Also, the proposed unmixing algorithm uses maximum likelihood method to estimate weight coefficients of Base Vectors which are used to represent mixed pixel. Finally, abundance estimation can be done using the new representation for endmembers and mixed pixel. The proposed algorithm is evaluated and compared with other state-of-the-art unmixing algorithms using simulated and real hyperspectral images. Experimental results demonstrate that the proposed unmixing algorithm can unmix pixels composed of correlated endmembers in hyperspectral images in the presence of spectral variability more accurately than previous methods.

Spectral mixture analysis (also called spectral unmixing) is one of the important and effective techniques to estimate abundance fractions of materials present in the hyperspectral imagery. Linear spectral mixture modeling is widely used in solving the spectral unmixing problems as its conciseness and clarity of physical meaning. This paper presents a novel algorithm to produce fully constrained (i.e. non-negative and sum to one constrained) abundance using the barycentric coordinates in the n-simplex. To impose non-negative constraint on the abundance, the proposed method use a serious of orthogonal projections to find the fully constrained solution, which takes into account the geometric structure of hyperspectral data set. The proposed algorithm is in line with the least squares criterion. The efficiency and effectiveness of the resulting unmixing algorithm is demonstrated using both synthetic and real hyperspectral images.

This paper investigates the use of Gaussian processes to detect non-linearly mixed pixels in hyperspectral images. The proposed technique is independent of nonlinear mixing mechanism, and therefore is not restricted to any prescribed nonlinear mixing model. The observed reflectances are estimated using both the least squares method and a Gaussian process. The fitting errors of the two approaches are combined in a test statistics for which it is possible to estimate a detection threshold given a required probability of false alarm. The proposed detector is compared to a robust nonlinearity detector recently proposed using synthetic data and is shown to provide a better detection performance. The new detector is also tested on a real hyperspectral image.

It is of great interest in spectral-spatial features classification for hyperspectral images (HSI) with high spatial resolution. This paper presents a novel Spectral-spatial classification method for improving hyperspectral image classification accuracy. Specifically, a new texture feature extraction algorithm exploits spatial texture feature from spectrum is proposed. It employs local binary patterns (LBPs) in order to extract the image texture feature with respect to spectrum information diversity (SID) to measure the differences of spectrum information. The classifier adopted in this work is support vector machine (SVM) because of its outstanding classification performances. In this paper, two real hyperspectral image datasets are used for testing the performance of the proposed method. Our experimental results from real hyperspectral images indicate that the proposed framework can enhance the classification accuracy compare to traditional alternatives.

The characteristics of SpecTIR and AVIRIS 16-bit hyperspectral images are analyzed. The requirements for compression of such images are formulated. The aspects of using the hierarchical compression algorithm in hyperspectral images storage are studied. Spectral component approximation algorithms are considered that allow both an increased compression ratio and retrieval of particular components. Interpolation algorithms are considered and a rank interpolator is offered for hyperspectral images compression. Real 16-bit hyperspectral images are used in computational experiments to investigate the efficiency of the proposed algorithms. The best parameters of these algorithms are found experimentally and general recommendations on how to tune the proposed hierarchical compression algorithm to suit hyperspectral images storage problems are given.