61 results about "Endmember" patented technology

A wavelet-based band difference-sum ratio method reduces the computation cost of classification and feature extraction (identification) tasks. A Generalized Difference Feature Index (GDFI), computed using wavelets such as Daubechies wavelets, is employed in a method to automatically generate a large sequence of generalized band ratio images. In select embodiments of the present invention, judicious data mining of the large set of GDFI bands produces a small subset of GDFI bands suitable to identify specific Terrain Category / Classification (TERCAT) features. Other wavelets, such as Vaidyanathan, Coiflet, Beylkin, and Symmlet and the like may be employed in select embodiments. The classification and feature extraction (identification) performance of the band ratio method of the present invention is comparable to that obtained with the same or similar data sets using much more sophisticated methods such as discriminants, neural net classification, endmember Gibbs-based partitioning, and genetic algorithms.

Target detection using algorithms is based on the adaptive real-time endmembers selection from hyperspectral image data. New pixels are correlated with prestored clutter endmembers to obtain subsequently detected target and clutter pixels and then suppressing the latter, leaving about 10% of the image as targets. Targets are then reduced in number by size filtering.

A method that facilitates identification of features in a scene enables enhanced detail to be displayed. One embodiment incorporates a multi-grid Gibbs-based algorithm to partition sets of endmembers of an image into smaller sets upon which spatial consistency is imposed. At each site within an imaged scene, not necessarily a site entirely within one of the small sets, the parameters of a linear mixture model are estimated based on the small set of endmembers in the partition associated with that site. An, enhanced spectral mixing process (SMP) is then computed. One embodiment employs a simulated annealing method of partitioning hyperspectral imagery, initialized by a supervised classification method to provide spatially smooth class labeling for terrain mapping applications. One estimate of the model is a Gibbs distribution defined over a symmetric spatial neighborhood system that is based on an energy function characterizing spectral disparities in both Euclidean distance and spectral angle.

A fast and rigorous multivariate curve resolution (MCR) algorithm is applied to remotely sensed spectral data. The algorithm is applicable in the solar-reflective spectral region, comprising the visible to the shortwave infrared (ranging from approximately 0.4 to 2.5 μm), midwave infrared, and thermal emission spectral region, comprising the thermal infrared (ranging from approximately 8 to 15 μm). For example, employing minimal a priori knowledge, notably non-negativity constraints on the extracted endmember profiles and a constant abundance constraint for the atmospheric upwelling component, MCR can be used to successfully compensate thermal infrared hyperspectral images for atmospheric upwelling and, thereby, transmittance effects. Further, MCR can accurately estimate the relative spectral absorption coefficients and thermal contrast distribution of a gas plume component near the minimum detectable quantity.

Described herein is a process for objectively and automatically determining spectral endmembers and transforming Spectral Mixture Analysis (SMA) from a widely used research technique into a user-friendly tool that can support the needs of all types of remote sensing. The process extracts endmembers from a spectral dataset using a knowledge-based approach. The process identifies a series of starting spectra that are consistent with a scene and its environment. The process then finds endmembers iteratively, selecting each new endmember based on a combination of physically and statistically-based tests. The tests combine spectral and spatial criteria and decision trees to ensure that the resulting endmembers are physically representative of the scene.

The disclosure concerns processing of electronic images, such as hyperspectral or multispectral images. In particular, but is not limited to methods, software and computer systems for determining underlying spectra of an image of a scene. The image data comprises for each pixel location a sampled image spectrum that is a mixture of plural reflectance spectra. Processor 310 determines or accesses plural hyperplanes that each have plural linearly independent basis vectors. Each hyperplane represents an estimate of one of the plural reflectance spectra. The processor 310 then determines for each pixel location, a contribution of the plural basis vectors of each hyperplane to the image spectrum of that pixel location. The processor 310 determines or accesses plural hyperplanes and not plural endmembers directly. Hyperplanes are two-dimensional while endmembers are only one-dimensional. As a result, hyperplanes carry more information, such as the illumination spectrum and are therefore of greater use. The decomposition of the image data into hyperplanes is possible without knowing the illumination spectrum. This decomposition allows for a range of new applications such as endmember extraction and compact representation.

According to one embodiment, an endmember spectrum extraction unit generates an input endmember spectrum group from a hyperspectral image. A collation unit outputs an endmember spectrum group including endmember spectra that are registered with neither a spectrum library nor an endmember spectrum database as a new endmember spectrum group. An input endmember spectrum distribution image generation unit generates an input endmember distribution image every input endmember spectrum. A new endmember spectrum distribution image selection unit selects an input endmember spectrum distribution image that is included in input endmember spectrum distribution images and that corresponds to new endmember spectrum as a new endmember distribution image. The user recognizes what kind of object corresponds to each new endmember spectrum on the basis of the new endmember spectrum distribution image, the hyperspectral image and a high resolution image, and determines and orders whether to register each new endmember spectrum with the endmember spectrum database.

A method is disclosed herein for compressing spectral data corresponding to an image comprising a plurality of pixels. The method includes the step of creating a set of potential endmembers. A first plurality of the potential endmembers are identified as a first set of endmembers based upon their respective correlations with a first spectral signature of a first of the plurality of pixels. The first pixel is then represented as a combination of the first set of endmembers. Processing of the image preferably continues by identifying a second plurality of the potential endmembers as a second set of endmembers based upon their respective correlations with a second spectral signature of a second of the plurality of pixels. The second pixel is then represented as a combination of the second set of endmembers.

Described herein is a process for objectively and automatically determining spectral endmembers and transforming Spectral Mixture Analysis (SMA) from a widely used research technique into a user-friendly tool that can support the needs of all types of remote sensing. The process extracts endmembers from a spectral dataset using a knowledge-based approach. The process identifies a series of starting spectra that are consistent with a scene and its environment. The process then finds endmembers iteratively, selecting each new endmember based on a combination of physically and statistically-based tests. The tests combine spectral and spatial criteria and decision trees to ensure that the resulting endmembers are physically representative of the scene.

The invention provides a method for identifying one or more materials in a scene by determining a set of spectral vectors, called endmembers, from a data set comprised of spectra from the image data, and matching the set of endmembers to predefined library materials. The image data of the scene is captured with a sensor, and comprises a plurality of spectra. The method applies an iterative mathematical criterion, termed residual minimization, to find the endmembers. The first endmember may be selected based on the largest mean square value or the largest mean magnitude value. Subsequent endmembers are determined by calculating weighting factors, such that the weighting factors are non-negative and the calculated vector differences, or residuals, generate the smallest error metric. The error metric is dependent upon the vector difference between two spectra in the image data set, and may be the mean squared vector difference between two spectra.

A method of identifying endmember spectral values from multispectral image data, where each multispectral data value is equal to a sum of mixing proportions of each endmember spectrum. The method comprises the steps of processing the data to obtain a multidimensional simplex having a number of vertices equal to the number of endmembers. The position of each vertex represents a spectrum of one of the endmembers. Processing the data is conducted by providing starting estimates of each endmember spectrum for each image data value. The mixing proportions for each data value is estimated from estimates of the spectra of all the endmembers. The spectrum of each endmember is estimated from estimates of the mixing proportions of the spectra of all the endmembers for each image data value. The estimation steps are repeated until a relative change in the regularised residual sum of squares is sufficiently small. The regularised residual sum of squares includes a term which is a measure of the size of the simplex.

A method of representing spectral data, such as hyperspectral imaging data (HSI) and multispectral imaging data (MSI), as a set of simplex models. The method finds end-images or end-spectra in the data (termed “endmembers”) as extreme points, and simultaneously determines the abundance of the endmembers.

A spectral unmixing scheme based on abundance sparse and endmember orthogonal constraint NMF(Non-negative Matrix Factorization) is an algorithm of a spectral image decomposition field. A classical NMFtarget function is a non-convex function, and a constraint condition needs to be added into the target function for solving the problems. The characteristics of spectral images are combined to put forward the spectral image unmixing algorithm which combines endmember orthogonality with abundance sparse constrains NMF on the basis of a linear spectral mixture model. The endmember orthogonal constraint guarantees independence among spectral endmembers, meanwhile, the abundance sparsity fully utilizes the sparsity of spectral data, two constraint conditions are introduced into the target function, a least square method is adopted to obtain the iteration rules of an endmember matrix and an abundance matrix, and a final result is obtained through the setting of an iterative termination condition. Through the experiments of simulation data and truthful data, the effectiveness of the algorithm is verified.

A downscaling method for a soil water content product is disclosed. The method includes A) acquiring passive microwave soil water content products of a region to be researched and optical remote sensing data at the same time, B) subjecting the optical remote sensing data to soil spectrum extraction based on a multiple endmember spectral mixture analysis process, C) building a model of a quantitative relation between soil spectrum reflection characteristics and soil water contents acquired from the passive microwave soil water content products by utilizing GA-PLS, and D) based on the model of the quantitative relation, building a soil water content downscaling model by utilizing the Taylor series expansion and acquiring soil water content data having a high spatial resolution. Advantages of passive microwave remote sensing data and optical remote sensing data on the temporal-spatial resolution are utilized in the method, and the passive microwave remote sensing data and the optical remote sensing data are effectively combined to acquire the soil water content data having the high spatial resolution, and therefore large-scale watershed scale region researching can be met, real-time or quasi real-time dynamic monitoring on watershed scale soil water contents is achieved, accuracy is high, building is easy and time and labor are saved.

The invention relates to a sparse unmixing method based on optical calculation. The method comprises the following steps: S1, an optical filtering method is adopted to acquire a transform domain sparse coefficient matrix for hyperspectral signals; S2, a deterministic measurement matrix is designed to realize perception measurement on the hyperspectral signals; S3, parallel multiplication calculation is carried out on the sparse coefficient matrix and the deterministic measurement matrix through a hardware FPGA for acquiring endmember signals; and S4, the sparse structural information of the hyperspectral signals and a structured reconstruction algorithm are used for reconstructing an abundance matrix. The optical filtering technology and an electrical system are adopted to realize multiscale geometric analysis, the sparse unmixing efficiency is greatly improved on the premise of ensuring the reconstruction precision, and quick and effective sparse decomposition can be realized.

Provided herein are algorithms and processes to extract endmembers from hyperspectral image data in real time. A Simplex Growing Algorithm is effective to estimate a p number of endmembers to be generated, to select one or more initial endmembers as a simplex of k members and to add a k+1 endmember to the simplex that yields a maximum simplex volume until k=p, thereby extracting one or more endmembers from the data. Alternatively, N-FINDR algorithms form an initial simplex set of p endmembers obtained from the hyperspectral image data, find a maximum volume of one or more initial p endmembers therewithin, replace one or more of the p endmembers within the simplex with one or more of the found p endmembers of maximum volume, and refind a maximum volume of p endmember(s) and replace p endmember(s) until no increase in p endmember(s) volume is found.

Described herein is a process for objectively and automatically determining spectral endmembers and transforming Spectral Mixture Analysis (SMA) from a widely used research technique into a user-friendly tool that can support the needs of all types of remote sensing. The process extracts endmembers from a spectral dataset using a knowledge-based approach. The process identifies a series of starting spectra that are consistent with a scene and its environment. The process then finds endmembers iteratively, selecting each new endmember based on a combination of physically and statistically-based tests. The tests combine spectral and spatial criteria and decision trees to ensure that the resulting endmembers are physically representative of the scene.

A method for rapid processing of large sets of hyperspectral data. A hyperspectral image, with hundreds of thousands to millions of pixels measured at hundreds of wavelengths, can contain over a gigabyte of data. Even modern computers can be quite slow when performing involved calculations on data sets of this size. An algorithm requiring a minimal amount of floating point calculations that still yielded useful results is disclosed.

The disclosure concerns processing of electronic images, such as hyperspectral or multispectral images. In particular, but is not limited to methods, software and computer systems for determining underlying spectra of an image of a scene. The image data comprises for each pixel location a sampled image spectrum that is a mixture of plural reflectance spectra. Processor 310 determines or accesses plural hyperplanes that each have plural linearly independent basis vectors. Each hyperplane represents an estimate of one of the plural reflectance spectra. The processor 310 then determines for each pixel location, a contribution of the plural basis vectors of each hyperplane to the image spectrum of that pixel location. The processor 310 determines or accesses plural hyperplanes and not plural endmembers directly. Hyperplanes are two-dimensional while endmembers are only one-dimensional. As a result, hyperplanes carry more information, such as the illumination spectrum and are therefore of greater use. The decomposition of the image data into hyperplanes is possible without knowing the illumination spectrum. This decomposition allows for a range of new applications such as endmember extraction and compact representation.

The invention discloses a latent fingerprint image enhancement method for hyperspectral imaging. The method comprises the following steps: step 1, acquiring a fingerprint endmember spectrum set; step 2, acquiring a background endmember spectrum set based on the fingerprint endmember spectrum set in the step 1 or combination of the fingerprint endmember spectrum set in the step 1 and average spectrum of different background regions marked with a user, and adopting orthogonal subspace iteration for extracting the background endmember spectrum set; step 3, separating fingerprints from backgrounds to realize fingerprint enhancement; and step 4, postprocessing fingerprint features: adopting morphological open-close mixed reconstruction for performing morphology-reserving and noise-inhibiting post-processing on the fingerprint features after enhancement in the step 3. The method is simple and practical, greatly lowers the difficulty of user operation, can smooth noises and fill holes on the premise of keeping enhanced fingerprint information, and is high in fingerprint enhancement reliability, high in efficiency and high in practicality and feasibility.

A method of identifying endmember spectral values from multispectral image data, where each multispectral data value is equal to a sum of mixing proportions of each endmember spectrum. The method comprises the steps of processing the data to obtain a multidimensional simplex having a number of vertices equal to the number of endmembers. The position of each vertex represents a spectrum of one of the endmembers. Processing the data is conducted by providing starting estimates of each endmember spectrum for each image data value. The mixing proportions for each data value is estimated from estimates of the spectra of all the endmembers. The spectrum of each endmember is estimated from estimates of the mixing proportions of the spectra of all the endmembers for each image data value. The estimation steps are repeated until a relative change in the regularised residual sum of squares is sufficiently small. The regularised residual sum of squares includes a term which is a measure of the size of the simplex.

Provided is a high-quality spectrum reconstruction method based on a few-wave-band high-resolution image and a low-resolution hyperspectral image. The method comprises the steps that the high-resolution image having few wave bands and the low-resolution hyperspectral image under the same scene are obtained; the low-resolution hyperspectral image is trained in a non-decomposition mode to obtain a spectrum dictionary; then, the high-resolution image having the few wave bands is subjected to sparse representation under the condition of being free of non-negative restriction, and a sparse representation coefficient is obtained; the part which cannot be represented by a sparse representation framework is estimated through spatial structure information; the dictionary, the sparse coefficient and the estimation part are utilized for reconstructing the hyperspectral image with high resolution. Accordingly, under the sparse representation framework, the non-decomposition mode is introduced to solve the spectrum dictionary, the deficiency of descriptive power of endmember decomposition to spectral characteristics of the hyperspectral image is made up, and the reconstructed spectrum precision, the effectiveness of hyperspectral image reconstruction and the spatial accuracy of reconstructing the hyperspectral image are effectively improved.

The invention relates to the technical field of remote sensing, and particularly relates to a thermal infrared fusion reconstruction method based on a hyperspectral image. The method comprises the following steps: A, pure endmembers are selected based on spectral reflectance characteristics and thermal response characteristics of an urban material; B, a multi-endmember spectral mixed analysis method is used to invert accurate terrain spatial abundance information; and C, through the surface temperature of joint representative endmembers and the spatial abundance information of the endmembers,a surface temperature image with a high spatial resolution is acquired. In the technical scheme of the invention, through a data fusion means, based on a hyperspectral remote sensing data reconstruction theory, the urban surface temperature image with a high spatial resolution is obtained, the image can significantly improve the spatial resolution of the urban temperature, and research on a seriesof problems brought about by an urban heat island and global changes can be done conveniently.

The invention relates to an automated morphological endmember extraction based hyperspectral image data unmixing method and belongs to the technical field of image data processing. According to the method, calculation is performed by utilizing a reference spectral vector to obtain an image element with highest mixing degree in a whole image, and based on this, a morphological operator is improved; an original structural element is improved with a new MEI calculation method; a structural element of an odd size is replaced with a structural element of an even size; and four candidate endmembers are selected from each structural element, so that possible information loss is effectively avoided, the accuracy of data processing is further improved, and finally the definition of the image obtained after data processing is improved; and the automated morphological endmember extraction based hyperspectral image data unmixing method is quite simple and convenient to implement.

According to one embodiment, an endmember spectrum extraction unit generates an input endmember spectrum group from a hyperspectral image. A collation unit outputs an endmember spectrum group including endmember spectra that are registered with neither a spectrum library nor an endmember spectrum database as a new endmember spectrum group. An input endmember spectrum distribution image generation unit generates an input endmember distribution image every input endmember spectrum. A new endmember spectrum distribution image selection unit selects an input endmember spectrum distribution image that is included in input endmember spectrum distribution images and that corresponds to new endmember spectrum as a new endmember distribution image. The user recognizes what kind of object corresponds to each new endmember spectrum on the basis of the new endmember spectrum distribution image, the hyperspectral image and a high resolution image, and determines and orders whether to register each new endmember spectrum with the endmember spectrum database.

The invention relates to the technical field of artificial intelligence and data processing, in particular to an automatic rock stratum identification method and system based on multispectral remote sensing, and the method specifically comprises the following steps: employing a continuum removal-based adaptive normalization method for a collected multispectral remote sensing data sample; decomposing the normalized sample based on a mixed pixel spectrum of a physical mechanism of a rock mass, and calculating an abundance coefficient of each end member; calculating a first derivative value of each normalized sample in each wave band, and calculating a fusion feature value of each dimension by combining an abundance coefficient; constructing a rock stratum identification model based on deep learning, and inputting fusion feature values of all dimensions to carry out rock stratum category probability prediction; performing optimization training on the rock stratum identification model; newly collected data are processed and then input into the trained model, and a final recognition result is obtained. The accuracy, the stability and the geological rationality of the automatic rock stratum identification result can be improved.

Provided herein are algorithms and processes to extract endmembers from hyperspectral image data in real time. A Simplex Growing Algorithm is effective to estimate a p number of endmembers to be generated, to select one or more initial endmembers as a simplex of k members and to add a k+1 endmember to the simplex that yields a maximum simplex volume until k=p, thereby extracting one or more endmembers from the data. Alternatively, N-FINDR algorithms form an initial simplex set of p endmembers obtained from the hyperspectral image data, find a maximum volume of one or more initial p endmembers therewithin, replace one or more of the p endmembers within the simplex with one or more of the found p endmembers of maximum volume, and refind a maximum volume of p endmember(s) and replace p endmember(s) until no increase in p endmember(s) volume is found.

The invention discloses an improved hyperspectral end member extraction method based on discrete variables, and the method comprises the steps of firstly, taking the minimum root-mean-square error between unmixed spectral data and original spectral data as an objective function, and building an end member extraction optimization model in combination with a constraint condition; calculating the food source information based on an artificial bee colony algorithm, selecting NP pieces of food source information as initial solutions and inputting the initial solutions into a genetic algorithm; andfinally, obtaining an optimal solution of the end member extraction optimization model based on a genetic algorithm. According to the method, the end member extraction precision optimization in the hyperspectral image is realized by applying the discrete artificial bee colony algorithm and the genetic algorithm to the hyperspectral end member extraction, so that the problems that a traditional endmember extraction method is low in precision and a linear constraint model is liable to fail are solved.

The invention discloses a hydrous mineral detection method and system based on the hyperspectrum of a Mars orbiter, and relates to the technical field of hyperspectral data processing and analys.The method comprises the steps that a distribution area of hydrous minerals is obtained based on Mars hyperspectral data; performing automatic end member extraction on the distribution area to obtain an image end member; and performing spectral feature inspection on the image end members, and performing unmixing based on an image end member library to obtain a hydrous mineral abundance map. According to the method, the characteristics of the water-containing minerals can be automatically identified, the mineral distribution range can be accurately positioned, meanwhile, the de-mixing end member library can be automatically constructed, the end members can be checked, and manual participation is reduced; the end member library is automatically constructed from the image based on the qualitative recognition result, and through overlay analysis and comparison with other unmixing end member construction methods, it is shown that the method has high credibility in the aspect of reflecting the distribution of the hydrous minerals and the abundance of the minerals.

The invention provides a hyperspectral image emissivity spectrum extraction method based on endmember abundance inversion. First, the whole infrared hyperspectral image is decomposed into mixed pixels to obtain the pure endmembers in the image and the corresponding endmembers. The abundance value of the target, and then obtain the radiance spectrum curve of each pixel point in the image, and then perform curve integration on the radiance spectrum of the target of interest within the set temperature threshold range, and estimate the infrared hyperspectral emissivity according to the integrated area The absolute black body temperature simulated by the curve, and then the emissivity spectrum curve is obtained. When extracting the radiance curve of the target of interest, the present invention considers the influence of mixed pixels, obtains the real radiance curve of the target through end-member extraction and abundance inversion, improves the extraction accuracy of the radiance curve, and then makes the emissivity The extraction accuracy of curves has been improved.