Search

This letter presents a novel approach for simultaneous depth and spectral imaging with a cross-modal stereo system. Two images of the target scene are captured at the same time: one compressively sampled hyperspectral measurement and one panchromatic measurement. The underlying hyperspectral cube is first reconstructed by leveraging the compressive sensing theory, during which a self-adaptive dictionary is learned from the panchromatic measurement to facilitate the reconstruction. The depth information of the scene is then recovered by estimating a disparity map between the hyperspectral cube and the panchromatic measurement through stereo matching. This disparity map, once obtained, is used to align the hyperspectral and panchromatic measurements to boost the hyperspectral reconstruction in an iterative manner. Through hardware experiments, for the first time to our knowledge, we demonstrate a snapshot system that allows for simultaneous depth and spectral imaging. The proposed system is capable of recording depth and spectral videos of dynamic scenes.

Compared to studies of analyzing hyperspectral image (HSI), finding spectral rectification, especially the seriously distorted pixels in boundaries, was seldom addressed in a clear way, albeit of first importance in HSI analysis and interpretation. In this paper, we present a simple but promising rectification method works in a coarse-to-fine framework for removing noise and enhancing useful features. Our approach, called CSR (coarse-to-fine spectral rectification), combines the theory of scale-aware with local smoothness for HSI rectification problem that is seldom pointed out. The useless information like noise in small scale is removed firstly. Then, the distinctive information like boundary in large scale is enhanced. The experimental result enjoys a built-in smoothing effect and a fact of the identical materials with same or similar signatures, which is suited for HSI subsequent application. Furthermore, our approach has powerful influence on both classification via five classifiers in terms of class labeled data and unmixing regarding to class unlabeled data. Rectified by a coarse-to-fine framework, our method presents superior performance and runs much faster than the competing methods.

A light field camera has a powerful and unique capability of light-ray collection and post-processing of data. In this paper, we describe a micro-lens-array-based approach to acquire the light field, with the added utility provided by simultaneous placement of linear variable filters (LVF) directly in the camera’s pupil plane. The new imaging spectrometer is capable of collecting the spectral segment information in the entire spectral range through only one exposure. Owing to the potential application of spectral detection, interest in the spectrum detection ability of the system is on the rise. As pixel sizes are reduced, consideration of diffraction effects in the system becomes increasingly important. We discuss a light field spectral imaging system and its wave propagation analysis. Then, we simulate a system response using our diffraction model and discuss the spectral resolution calibration with the application of the system model. In addition, the spectral resolution of the light field spectral imaging system is tested using the monochromator method; the results are found to be in good agreement with the simulation results.

Mapping forests is an important process in managing natural resources. At present, due to spectral resolution limitations, multispectral images do not give a complete separation between different forest species. In contrast, advances in remote sensing technologies have provided hyperspectral tools and images as a solution for the determination of species. In this study, spectral signatures for stone pine (Pinus pinea L.) forests were collected using an advanced spectroradiometer “ASD FieldSpec 4 Hi-Res” with an accuracy of 1 nm. These spectral signatures are used to compare between different multispectral and hyperspectral satellite images. The comparison is based on processing satellite images: hyperspectral Hyperion, hyperspectral CHRIS-Proba, Advanced Land Imager (ALI), and Landsat 8. Enhancement and classification methods for hyperspectral and multispectral images are investigated and analyzed. In addition, a well-known hyperspectral image classification algorithm, spectral angle mapper (SAM), has been improved to perform the classification process efficiently based on collected spectral signatures. The results show that the modified SAM is 9% more accurate than the conventional SAM. In addition, experiments indicate that the CHRIS-Proba image is more accurate than Landsat 8 (overall accuracy 82%, precision 93%, and Kappa coefficient 0.43 compared to 60, 67%, and 0.035, respectively). Similarly, Hyperion is better than ALI in mapping stone pine (overall accuracy 92%, precision 97%, and Kappa coefficient 0.74 compared to 52, 56%, and − 0.032, respectively).

Space-borne hyperspectral imagery is an important tool for earth sciences and industrial applications. Higher spatial and spectral resolutions have been sought persistently, although this results in more power, larger volume and weight during a space-borne spectral imager design. For miniaturization of hyperspectral imager and optimization of spectral splitting methods, several methods are compared in this paper. Spectral time delay integration (TDI) method with high transmittance Integrated Stepwise Filter (ISF) is proposed.With the method, an ISF imaging spectrometer with TDI could achieve higher system sensitivity than the traditional prism/grating imaging spectrometer. In addition, the ISF imaging spectrometer performs well in suppressing infrared background radiation produced by instrument. A compact shortwave infrared (SWIR) hyperspectral imager prototype based on HgCdTe covering the spectral range of 2.0–2.5 μm with 6 TDI stages was designed and integrated. To investigate the performance of ISF spectrometer, a method to derive the optimal blocking band curve of the ISF is introduced, along with known error characteristics. To assess spectral performance of the ISF system, a new spectral calibration based on blackbody radiation with temperature scanning is proposed. The results of the imaging experiment showed the merits of ISF. ISF has great application prospects in the field of high sensitivity and high resolution space-borne hyperspectral imagery.

Spectral imaging involves the sensing of a large amount of spatial information across a multitude of wavelengths. Conventional approaches rely on scanning techniques to construct a spectral data cube. Recently, compressive spectral imaging (CSI) has allowed to estimate spectral images with as few as a single coded snapshot. On a different front, 3-D ranging imaging often involves scanning along one of the spatial dimensions to estimate the depth of an scene using structured light, or the use of two cameras as required by stereo-imaging techniques. Recently, Time-of-Flight (ToF) snapshot imaging has gained considerable attention, due to its accuracy and speed. To date, however, these imaging modalities (CSI and ToF) have been realized and implemented by separate independent imaging sensors. This paper presents the development of a single aperture compressive spectral + depth imaging camera that employs a commodity 3-D range ToF sensor as the sensing device of a coded-aperture-based compressive spectral imager. The proposed system uses a single aperture/single sensor; thus, representing a significant improvement over existing RGB+D cameras that integrate two separate image sensors, one for RGB and another for depth. In addition, the observable wavelength range of the CSI device is expanded from the visible to the near-infrared, resolving up to as many as 16 independent channels. The proposed system allows the addition of side-information by placing a grayscale or RGB camera in the same path of the single-aperture system, such that the quality of the spectral estimation is improved, while maintaining high-frame rates. We demonstrate the proposed ideas through real experimentation conducted on an assembled CSI+ToF testbed camera system.

A hyperspectral image (HSI) super-resolution (SR) is a highly attractive topic in computer vision. However, most existed methods require an auxiliary high-resolution (HR) image with respect to the input low-resolution (LR) HSI. This limits the practicability of these HSI SR methods. Moreover, these methods often destroy the important spectral information. This letter presents a deep spectral difference convolutional neural network (SDCNN) with the combination of a spatial-error-correction (SEC) model for HSI SR. This method allows for full exploration of the spectral and spatial correlations, which achieves a good spatial information enhancement and spectral information preservation. In the proposed method, the key band is automatically selected and super-resolved with the boundary bands. Meanwhile, spectral difference mapping between the LR and HR HSIs can be learned by the SDCNN, and then be transformed according to the SEC model, which aims at correcting the spatial error while preserving the spectral information. The rest nonkey bands will be super-resolved under the guidance of the transformed spectral difference. Experimental results on synthesized and real-scenario HSIs suggest that the proposed method: 1) achieves comparable performance without requiring any auxiliary images of the same scene and 2) requires less computation time than the state-of-the-art methods.

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.

Multispectral image contains much more information than regular RGB image so that it has wide applications in the field of physics, surveillance, environment monitoring or investigation of artworks. The traditional spectral imaging cameras are mainly based on broom/scanning or filter wheel imaging technologies, while those techniques share significant disadvantages of high cost and structural complexity. In this paper we proposed an eight-band multispectral camera prototype with only one CCD, which is cost-effective and high-speed for capturing the moving objects. A simulated experiment is designed to evaluate the proposed eight-band imaging system. The hyperspectral scene is firstly acquired and imaged on the CCD which generates the raw image, and then a demosaicking algorithm based on modified bilinear interpolation is employed to reconstruct the missing imaging values from the raw image, at last the demosaicked image is converted into spectral values with the conversion coefficient calculated from spectral camera characterization. In order to evaluate the proposed eight-band imaging system, the demosaicked image values and recovered spectral values are compared with the original data respectively in the form of PSNR and RRMS. In the experiment result, the eight-band multispectral camera captures spectral images successfully with high PSNR values and lower RRMS errors.

Low-cost flight-based hyperspectral imaging systems have the potential to provide important information for ecosystem and environmental studies as well as aide in land management. To realize this potential, methods must be developed to provide large-area surface reflectance data allowing for temporal data sets at the mesoscale. This paper describes a bootstrap method of producing a large-area, radiometrically referenced hyperspectral data set using the Landsat surface reflectance (LaSRC) data product as a reference target. The bootstrap method uses standard hyperspectral processing techniques that are extended to remove uneven illumination conditions between flight passes, allowing for radiometrically self-consistent data after mosaicking. Through selective spectral and spatial resampling, LaSRC data are used as a radiometric reference target. Advantages of the bootstrap method include the need for minimal site access, no ancillary instrumentation, and automated data processing. Data from two hyperspectral flights over the same managed agricultural and unmanaged range land covering approximately 5.8 km2 acquired on June 21, 2014 and June 24, 2015 are presented. Data from a flight over agricultural land collected on June 6, 2016 are compared with concurrently collected ground-based reflectance spectra as a means of validation.

In recent years many optoelectronic techniques have been developed for improvement and the development of devices for tissue analysis. Spectral-Domain Optical Coherence Tomography (SD-OCT) is a new medical interferometric imaging modality that provides depth resolved tissue structure information with resolution in the μm range. However, SD-OCT has its own limitations and cannot offer the biochemical information of the tissue. These data can be obtained with hyperspectral imaging, a non-invasive, sensitive and real time technique. In the present study we have combined Spectral-Domain Optical Coherence Tomography (SD-OCT) with Hyperspectral imaging (HSI) for tissue analysis. The Spectral-Domain Optical Coherence Tomography (SD-OCT) and Hyperspectral imaging (HSI) are two methods that have demonstrated significant potential in this context. Preliminary results using different tissue have highlighted the capabilities of this technique of combinations.

In order to provide reliable information about the instantaneous field of view considered in hyperspectral images through spectral unmixing, understanding the kind of mixture that occurs over each pixel plays a crucial role. In this paper, in order to detect nonlinear mixtures, a method for fast identification of linear mixtures is introduced. The proposed method does not need statistical information and performs an a priori test on the spectral linearity of each pixel. It uses standard least squares optimization to achieve estimates of the likelihood of occurrence of linear combinations of endmembers by taking advantage of the geometrical properties of hyperspectral signatures. Experimental results on both real and synthetic data sets show that the aforesaid algorithm is actually able to deliver a reliable and thorough assessment of the kind of mixtures present in the pixels of the scene.

This letter presents a novel mixed noise (i.e., Gaussian, impulse, stripe noises, or dead lines) reduction method for hyperspectral image (HSI) by utilizing low-rank representation (LRR) on spectral difference image. The proposed method is based on the assumption that all spectra in the spectral difference space of HSI lie in the same low-rank subspace. The LRR on the spectral difference space was exploited by nuclear norm of difference image along the spectral dimension. It showed great potential in removing structured sparse noise (e.g., stripes or dead lines located at the same place of each band) and heavy Gaussian noise. To simultaneously solve the proposed model and reduce computational load, alternating direction method of multipliers was utilized to achieve robust reconstruction. The experimental results on both simulated and real HSI data sets validated that the proposed method outperformed many state-of-the-art methods in terms of quantitative assessment and visual quality.

This paper presents an interactive feedback scheme of spatial resolution enhancement and spectral unmixing in hyperspectral imaging. Traditionally spatial resolution enhancement and spectral unmixing operations have been carried out separately, often in series. In such sequential processing, spatially enhanced hyperspectral images (HSIs) may introduce distortion in spectral fidelity making spectral unmixing results unreliable, or vice versa. Since both high- and low-resolution HSIs have the same endmembers, the deviation in spectral unmixing between targets and estimated high-resolution HSIs can be used as feedback to control spatial resolution enhancement. The spatial difference before and after unmixing can also be used as feedback to enhance spectral unmixing. Therefore, spectral unmixing is utilized as a constraint to spatial resolution enhancement, while spatial resolution enhancement helps improve spectral unmixing results. The performance of spatial resolution enhancement and spectral unmixing can be improved since one behaves like a prior to the other. Experimental results on both simulated and real HSI data sets demonstrate that the proposed interactive feedback scheme simultaneously achieved spatial resolution enhancement and spectral unmixing fidelity. This paper is an extended version of the previous work.

Over the past decade, anomaly detection has been an alluring topic in hyperspectral imagery. Despite most anomaly detection methods that focus on spectral information for detecting targets, this letter proposes a method in which both spectral and spatial information about the hyperspectral image (HSI) has been utilized for detecting anomalies. The differential attribute profile anomaly detection (DAPAD) method utilizes principal component analysis and DAP to extract spectral and spatial information from HSI, respectively. DAPs can model different kinds of structural information in a scene, which makes better extraction of spatial information. The proposed method is applied in experiments on Hyperspectral Digital Imagery Collection Experiment remote sensing data and the experimental results confirm the DAPAD method’s superiority over five commonly used state-of-the-art anomaly detection methods.

Hyperspectral images comprise hundreds of narrow contiguous wavelength bands which include wealth spectral information, and a great potential of Light detection and ranging (LIDAR) data lies in its benefits of height measurements, which can be used as complementary information for the classification of hyperspectral data. In this paper, a feature-fusion strategy of hyperspectral and LIDAR data is taken into account in order to develop a new classification framework for the accurate analysis of a surveyed area. The proposed approach employs extinction profiles (EPs) extracted with extinction filters computed on both hyperspectral and LIDAR images, leading to a fusion of the spectral, spatial, and elevation features. Experimental results obtained by using a real hyperspectral image along with LIDAR-derived digital surface model (DSM) collected over the University of Houston campus and its neighboring urban area demonstrate the effectiveness of the proposed framework.

Spectrally resolved fluorescence lifetime imaging1–3 and spatial multiplexing1,4,5 have offered information content and collection-efficiency boosts in microscopy, but efficient implementations for macroscopic applications are still lacking. An imaging platform based on time-resolved structured light and hyperspectral single-pixel detection has been developed to perform quantitative macroscopic fluorescence lifetime imaging (MFLI) over a large field of view (FOV) and multiple spectral bands simultaneously. The system makes use of three digital micromirror device (DMD)-based spatial light modulators (SLMs) to generate spatial optical bases and reconstruct N by N images over 16 spectral channels with a time-resolved capability (~40 ps temporal resolution) using fewer than N2 optical measurements. We demonstrate the potential of this new imaging platform by quantitatively imaging near-infrared (NIR) Förster resonance energy transfer (FRET) both in vitro and in vivo. The technique is well suited for quantitative hyperspectral lifetime imaging with a high sensitivity and paves the way for many important biomedical applications.

Colored coded aperture optimization in compressive spectral imaging is discussed. Based on the analysis of the coherence of the underlying sensing matrix, a general family of codes is derived. These designs lead to reconstructions of multispectral scenes of better quality than the ones obtained using the traditional random black and white coded apertures. The approach used in this work exploits the structure of the sensing matrices and reduces the problem of the design of a large-scale matrix to a subset of substantially smaller problems for which it is possible to obtain a closed form solution, leading to fast design algorithms.

Within-class spectral variation, which is caused by varied imaging conditions, such as changes in illumination, environmental, atmospheric, and temporal conditions, significantly degrades the performance of hyperspectral image classification. Recent studies have shown that such spectral variation can be alleviated by exploring the low-rank property in the spectral domain, especially based on the low-rank subspace assumption. In this paper, the low-rank subspace assumption is approached by exploring the low-rank property in the local spectral domain. In addition, the low-rank property in the spatial domain is also explored to alleviate spectral variation. As a result, two novel spectral-spatial low-rank (SSLR) strategies are designed to alleviate spectral variation by exploring the low-rank property in both spectral and spatial domains. Experimental results on two benchmark hyperspectral datasets demonstrate that exploring the low-rank property in local spectral space can help to alleviate spectral variation and improve the performance of classification obviously for all tested data, while exploring the low-rank property in spatial space is more effective for images presenting large homogeneous areas.

One way to increase the amount of information acquired via hyperspectral imaging and therefore to increase the possibility of data analysis is combining the spatial and spectral information of hyperspectral data sets. The aforementioned data sets are obtained by cameras covering different spectral ranges. The purpose of this article is to develop an algorithm which is able to combine two data sets acquired by two hyperspectral pushbroom imagers, covering the visible (VIS) and the near infrared (NIR) wavelength range. Initially, the effect of optical aberrations, as well as errors via the image registration were examined. Subsequently a correction algorithm for both the optical aberration and the image registration is elaborated.

Owning to the absorption of radiation by ground materials on specific wavelengths, spectral curve shapes stay relatively stable in hyperspectral images under various conditions and can be used for material identification. Given a reference spectrum, ground materials can be identified by matching the spectral curve shape of the reference spectrum to that of observed spectra. In the past decade, the spectral curve shape matching approach of the Tetracorder system (SCSMT) has been widely used for material identification. However, the SCSMT approach is performed directly on the continuum-removed spectra. It only catches the shape features on a large scale, missing the fine structures on spectral curves. To consider this issue, a new spectral shape matching approach is proposed, which employs the first-order and second-order spectral derivatives to capture the fine structures on spectra. A new metric is designed to balance the contributions of spectral curves, their first-order and second-order derivatives. To evaluate the performance of the proposed approach, it is applied to synthesized spectra as well as to identify oil spills from two practical hyperspectral images. The experimental results show that the proposed approach can improve the detection performance of the SCSMT approach.

The classification of hyperspectral images with a few labeled samples is a major challenge which is difficult to meet unless some spatial characteristics can be exploited. In this study, we proposed a novel spectral-spatial hyperspectral image classification method that exploited spatial autocorrelation of hyperspectral images. First, image segmentation is performed on the hyperspectral image to assign each pixel to a homogeneous region. Second, the visible and infrared bands of hyperspectral image are partitioned into multiple subsets of adjacent bands, and each subset is merged into one band. Recursive edge-preserving filtering is performed on each merged band which utilizes the spectral information of neighborhood pixels. Third, the resulting spectral and spatial feature band set is classified using the SVM classifier. Finally, bilateral filtering is performed to remove “salt-and-pepper” noise in the classification result. To preserve the spatial structure of hyperspectral image, edge-preserving filtering is applied independently before and after the classification process. Experimental results on different hyperspectral images prove that the proposed spectral-spatial classification approach is robust and offers more classification accuracy than state-of-the-art methods when the number of labeled samples is small.

Remotely sensed hyperspectral images exhibit very high dimensionality in the spectral domain. As opposed to band selection techniques, which extract a subset of the original spectral bands in the image, spectral partitioning (SP) techniques reassign the original bands into subgroups that are then processed separately. From a classification perspective, this strategy has the advantage that all the original information in the hyperspectral data can be retained while addressing the curse of dimensionality given by the Hughes phenomenon. Even if SP prior to classification has been widely used, the strategies adopted to perform such partitioning did not consider the diversity of spectral classes in the scene. In other words, available techniques are not driven by the information contained in the classes of interest, which can be very useful to perform the SP in a more effective manner for classification purposes. To address this issue, in this paper, we present a new class-oriented SP technique that exploits prior information about the classes by automatically ranking the spectral bands that are more useful for each specific class (instead of considering the hyperspectral image as a whole). The resulting multiple subgroups of bands with lower dimensionality are then fed to a multiple classifier system. Our experimental results, conducted with three different hyperspectral airborne images, suggest that the presented method leads to competitive results when compared to other state-of-the-art approaches in the field.

The efficiency of methods of the spectral-spatial classification of similarly looking types of vegetation on the basis of hyperspectral data of remote sensing of the Earth, which take into account local neighborhoods of analyzed image pixels, is experimentally studied. Algorithms that involve spatial pre-processing of the raw data and post-processing of pixel-based spectral classification maps are considered. Results obtained both for a large-size hyperspectral image and for its test fragment with different methods of training set construction are reported. The classification accuracy in all cases is estimated through comparisons of ground-truth data and classification maps formed by using the compared methods. The reasons for the differences in these estimates are discussed.

In hyperspectral images, spectral mixing occurs when objects lying beside each cannot be distinguished as different entities due to its low spatial resolution. Other hurdles in hyperspectral imaging are its huge dimension and noisy bands. In this paper, a new approach for spectral unmixing is presented where, the data is reduced dimensionally and, the bands eliminated during this are denoised using the existing denoising methods. Then, dataset with these bands is dimensionally reduced and their presence after reduction is validated using spectral unmixing methods. The effectiveness of this method is evaluated using parametric measures such as RMSE and classification accuracy.

A method is considered for rendering coastal water depths according to multi- and hyperspectral remote sensing imagery in the visible and near-infrared spectral regions. The depth is recovered for each pixel on the basis of solution of the inverse problem, which consists in artificial neural network learning with the use of a semianalytical model of radiation transfer in water, taking into account the effects of light scattering and absorption in the underwater light field, at least in three informative spectral channels for each bottom type. A possibility of adjusting the learning process is provided by the use of regression algorithms for determining organic and mineral impurities in water from their in-situ measurements. We enriched the library of the spectral characteristics of different bottom types and found informative identifiers for them. The results are tested on aircraft and hyperspectral space imagery data.

With nanometric spectral resolution and number of bands as high as 220, Hyper spectral sensors like Hyperion and AVIRIS are gaining wide appreciation. Narrow, continuous wavelength of bands upon a vast spectrum of electromagnetic wavelength enables better precision in identification of materials by distinguishing between their unique spectral signatures. However, their poor spatial resolution is a major impediment in realising the full potential of hyperspectral imaging. Efforts are being made worldwide to improve the spatial resolution of hyperspectral imagery by fusing them with registered panchromatic imagery of higher resolution. However, most of the conventional methods fail to preserve the spectral fidelity of the fused images due to severe color distortion. Here, we propose an efficient paradigm to sharpen hyperspectral imagery with high spatial information content and minimal color distortion using color normalization by Lαβ and intensity image generation using Spectral Mixture Analysis. Quantitative assessment indices have been calculated to prove that our method is superior in terms of minimization of color distortion and maximization of spatial details as compared to other methods.

The development and processing of three-dimensional images as a “hypercube” of spectral data in hyperspectral optical–electronic remote sensing systems are described in a formalized manner. The correlation identification of observed objects on the basis of spectral features is considered. The criterion for determining of similarity between vectors of recorded and reference spectral images of objects is based on their cross-correlation. Taking into the fact that the total spectral data array recorded by currently applicable hyperspectrometers is excessive for the solution of many issues related to remote sensing of the Earth, this paper proposes a method making it possible to reduce spectral data redundancy by selection of the most informative spectral channels. The essential dimension of the spectral data makes it possible to solve issues related to identification and classification of objects by spectral features through a limited number of very informative spectral channels selected in the areas where the function describing a spectral image of the observed object undergoes well-defined changes in behavior. The algorithm for selection of the most informative spectral channels, which is based on the determination of jump coordinates (major changes) of a spectral image, is substantiated. The selected channels meet the maximum likelihood criterion. The obtained experimental research data on object identification quality with involvement of real hyperspectral data of aerospace Earth remote sensing systems are reported. Five to twenty spectral readouts are needed to provide identification by a limited number of very informative spectral channels. This confirms the idea of existing essential dimensionality of the spectral data.

This letter proposes a new approach for the spectral–spatial classification of hyperspectral images, which is based on a novel extrema-oriented connected filtering technique, entitled as extended extinction profiles . The proposed approach progressively simplifies the first informative features extracted from hyperspectral data considering different attributes. Then, the classification approach is applied on two well-known hyperspectral data sets, i.e., Pavia University and Indian Pines, and compared with one of the most powerful filtering approaches in the literature, i.e., extended attribute profiles. Results indicate that the proposed approach is able to efficiently extract spatial information for the classification of hyperspectral images automatically and swiftly. In addition, an array-based node-oriented max-tree representation was carried out to efficiently implement the proposed approach.

Spectral unmixing pursues the identification of spectrally pure constituents, called endmembers , and their corresponding abundances in each pixel of a hyperspectral image. Most unmixing techniques have focused on the exploitation of spectral information alone. Recently, some techniques have been developed to take advantage of the complementary information provided by the spatial correlation of the pixels in the image. Computational complexity represents a major problem in these spatial–spectral techniques, as hyperspectral images contain very rich information in both the spatial and spectral domains. In this letter, we develop a computationally efficient implementation of a spatial–spectral processing algorithm that has been successfully applied prior to the spectral unmixing of the hyperspectral data. Our implementation has been optimized for the commodity graphics processing units (GPUs) and is evaluated (using both synthetic and real data) using different GPU architectures. Significant speedups can be achieved when processing hyperspectral images of different sizes. This allows for the inclusion of the proposed parallel preprocessing module in a full hyperspectral unmixing chain able to operate in real time.

Imaging spectroscopy is widely used in weed recognition, pest monitoring, agricultural product quality control and other precision agricultural fields. In the present study, an in-house-designed/developed field imaging spectroscopy system (FISS, 380–870 nm) was used to obtain the imaging spectra of soybean leaves at 344 wavelengths. The spatial and spectral information including the entropy, mean reflectivity and standard deviation of the leaf images at different wavelengths were extracted; the chlorophyll content was retrieved using multiple linear regression (MLR) together with the spatial information and spectral information, and the results were compared with the results derived with the Analytical Spectral Devices (ASD, FieldSpecFR spectrometer, Analytical Spectral Devices Inc., USA) data that were generated using conventional single sensor spectrometers. The results demonstrated that the entropy, standard deviation and other features of the image were very good indicators of the leaf chlorophyll content, confirming the idea that spatial information can be used to retrieve chlorophyll content, with an accuracy equivalent to that of spectral information, and can provide information that spectral reflectivity cannot provide. Thus, integrating spatial information and spectral information can greatly improve the chlorophyll content retrieval accuracy and reduce the estimation errors by 20 %. Due to the unique measurement method and image-spectrum-in-one feature, the field imaging spectroscopy system (FISS) data can be conveniently used to achieve accurate chlorophyll content retrieval, and the retrieval error was reduced by 30–45 % compared with that for the ASD data. FISS data and the proposed method of integrating both spectral and spatial information of imaging spectroscopy have potential advantages in quantitative spectral analysis applied in agricultural biochemistry related fields.

In recent years hyperspectral imaging has proved its significance in the detection and mapping of various objects of interest in a scene. Various methods for object detection in hyperspectral images have been developed with their advantages and limitations. In the present study, a methodology comprising spectral derivative (first order) and spectral information divergence has been investigated for detection of objects in hyperspectral images. The efficacy of the detection scheme has been examined over two different hyperspectral data sets of Hyperion images. Tea plants (Camellia sinensis) and Sal trees (Shorea robusta) (pure pixels) have been detected as the objects of interest in the hyperspectral images independently with reduced false pixels. The proposed methodology may in future be applied for classification of mixed pixels.

Anomalies usually refer to targets with a spot of pixels (even subpixels) that stand out from their neighboring background clutter pixels in hyperspectral imagery (HSI). Compared to backgrounds, anomalies have two main characteristics. One is the spectral anomaly, i.e., their spectral signatures are different from those associated to their surrounding backgrounds; another is the spatial anomaly, i.e., anomalies occur as few pixels (even subpixels) embedded in the local homogeneous backgrounds. However, most of the existing anomaly detection algorithms for HSI only employed the spectral anomaly. If the two characteristics are exploited in a detection method simultaneously, better performance may be achieved. The third-order (two modes for space and one mode for spectra) tensor representation of HSI has been proved to be an effective tool to describe the spatial and spectral information equivalently; therefore, tensor representation is convenient for exhibiting the two characteristics of anomalies simultaneously. In this paper, a new anomaly detection method based on tensor decomposition is proposed and divided into three steps. Three factor matrices and a core tensor are first estimated from the third-order tensor that is constructed from the HSI data cube by using the Tucker decomposition, and their major and minor principal components (PCs) are more likely to correspond to the spectral signatures of the backgrounds and the anomalies, respectively. In the second step, a reconstruction-error-based method is presented to find the first largest PCs along each mode to eliminate the spectral signatures of the backgrounds as much as possible, and thus, the remaining data may be modeled as the spectral signatures of the anomalies with a Gaussian noise. Finally, a CFAR test is implemented to detect the anomalies from the remaining data. Experiments with simulated, synthetic, and real HSI data sets reveal that the proposed method outperforms those spectral-anomaly-based methods with better detection probability and less false alarm rate.

This paper presents a novel spatial–spectral classification method for remotely sensed hyperspectral images. First of all, a multiscale representation technique based on random projection, referred as random multiscale representation (RMSR), is proposed to extract the spatial features from the given scene. The idea behind RMSR is to properly model the spatial characteristics comprised by each pixel vector and its neighbors by some criteria computed at all reasonable scales, and then compress the implicit high-dimensional spatial features by using a very sparse measurement matrix that approximately preserves the salient spatial information. The entire process is explicitly performed by computing simple criteria (i.e., the first two moments) at rectangular scales of random bands, according to the nonzero entries of the sparse measurement matrix. Subsequently, a composite kernel framework is utilized to balance the extracted spatial features and the original spectral features in the classifier. Our proposed method is shown to be effective for hyperspectral image classification purposes. Specifically, our experimental results with hyperspectral images collected by the airborne visible/infrared imaging spectrometer and the reflective optics spectrographic imaging system demonstrate the effectiveness of the proposed method as compared to other state-of-the-art spatial–spectral classifiers.

Due to the limitation of hyperspectral sensors and optical imaging systems, there are several irreconcilable conflicts between high spatial resolution and high spectral resolution of hyperspectral images (HSIs). Therefore, HSI super-resolution (SR) is regarded as an important preprocessing task for subsequent applications. In this letter, we use sparse representation to analyze the spectral and spatial feature of HSIs. Considering the sparse characteristic of spectral unmixing and high pattern repeatability of spatial–spectral blocks, we proposed a novel HSI SR framework utilizing spectral mixture analysis and spatial–spectral group sparsity. By simultaneously combining the sparsity and the nonlocal self-similarity of the images in the spatial and spectral domains, the method not only maintains the spectral consistency but also produces plenty of image details. Experiments on three hyperspectral data sets confirm that the proposed method is robust to noise and achieves better results than traditional methods.

In this paper, we assert that a recently developed approach to classify hyperspectral images provides substantial robustness to illumination effects for hyperspectral image analysis. The approach [angular discriminant analysis (ADA) and its variants] seeks to find lower dimensional subspaces, where class-specific information is angularly well-separated and where appropriate classifiers can then be applied. We demonstrate the ADA, and its local and kernel variants provide substantial robustness to data which is affected by such illumination differences (commonly encountered in geospatial images). We validated with two different hyperspectral datasets, both of which contain classes in well-illuminated regions and in shadows. We note that the proposed framework is suitable to the general problem of illumination variation in scenes—we validate with a specifically harsh and commonly encountered scenario where classes in an image are present in regions where they are well illuminated, and regions where they are occluded by shadows. Experimental results show that this dimensionality reduction not only enables robustness to sample size by virtue of feature reduction but also is very effective at capturing class-specific information irrespective of changes in illumination content, in the lower dimensional subspace.

In this paper, two novel multidimensional striping noise compensation (SNC) algorithms for push-broom hyperspectral cameras (PBHCs) have been developed. The SNC algorithms employ a novel pixelwise, affine image-degradation model, which assumes that the striping noise (SN) parameters are spatially uncorrelated, spectrally independent, and decoupled from the camera’s spectral response. Algorithms simultaneously exploit the spatial and temporal information contained in an image as well as the spectral information contained at adjacent spectral images. The multidimensional SNC algorithms were successfully tested on real hyperspectral data from both a commercial PBHC operating in the spectral range of 400–1000 nm, at a resolution of 1.04 nm, and the ESA earth-observing CHRIS/PROBA sensor.

This paper presents a novel panoramic hyperspectral stereo imaging system with depth map estimation capability. Images are collected from two line scan hyperspectral cameras in stereo configuration, while the platform is rotating. The proposed system combines the strengths of hyperspectral, stereo, and panoramic imaging. In addition, the proposed system uses passive sensors and hence does not need any radiation source, unlike LiDAR-based systems. In this paper, the proposed system, which captures hyperspectral data from 400 to 1000 nm wavelengths, is introduced, the stereo calibration procedure is described, corresponding analytic analyses are carried out with relation to rotating line-scan camera theory, and critical system parameters and aspects are addressed. In addition, a novel multiband stereo matching approach, which introduces multiband Census transform and SAD-based cost aggregation to stereo matching literature, is proposed. Experimental results are evaluated using two hyperspectral datasets, and the performance of the proposed algorithm is compared with the performance of several prominent matching algorithms available in the literature. A LiDAR-based system is used to enable an analysis of depth information accuracy. The experimental results show the enhanced matching and depth estimation performance of the proposed matching approach over standard stereo methods, and the consistency of depth values between LiDAR and the proposed system. The proposed system, which combines the strengths of panoramic imaging and stereo imaging with the high spectral resolution of hyperspectral cameras, uses passive imaging solutions and is fit to be used in applications requiring depth computation, target detection, and change detection.

This study proposes a methodology for simulating pan-sharpening using hyperspectral (HS) data to evaluate the influence of the spectral properties (band position and band width) of panchromatic (PAN) data on the quality of pan-sharpened data. PAN images were generated from Airborne Visible/Infrared Imaging Spectrometer HS data by changing the position and width of the spectral waveband, and these images were subjected to pan-sharpening with multispectral (MS) bands, also generated from HS data. The qualities of pan-sharpened MS data were evaluated using five measures: erreur relative globale adimensionnelle de synthese, weighted signal-to-noise ratio, universal image quality index, quality index based on local variance and objective image fusion performance measure. This methodology was applied to three pan-sharpening methods (Gram–Schmidt spectral sharpening, block-based synthetic variable ratio and generalised Laplacian pyramid with spectral distortion minimisation) to characterise the spectral ...

International audience; The Intrinsic Dimensionality (ID) of multivariate data is a very important concept in spectral unmixing of hyperspectral images. A good estimation of the ID is crucial for a correct retrieval of the number of endmembers (the spectral signatures of macroscopic materials) in the image, for dimensionality reduction or for subspace learning, among others. Recently, some approaches to perform spectral unmixing and super-resolution locally have been proposed, which require a local estimation of the number of endmembers to use. However, the role of ID in local regions of hyperspectral images has not been properly addressed. Some important issues when dealing with small regions of hyperspectral data can seriously affect the performance of conventional hyperspectral ID estimators. We show that three factors mainly affect local ID estimation: the number of pixels in the local regions, which has to be high enough for the estimations to be relevant, the number of hyperspectral bands which complicates the estimations if the ambient space has a high dimensionality, and the noise, which can be misinterpreted as signal when its power is important. Here, we review the hyperspectral ID estimators on the literature for local ID estimation, we show how they behave in a local setting on synthetic and real datasets, and we provide some guidelines to make proper use of these estimators in local approaches.

Hyperspectral sensors represent a powerful tool for chemical mapping of solid-state samples, since they provide spectral information localized in the image domain in very short times and without the need of sample pretreatment. However, due to the large data size of each hyperspectral image, data dimensionality reduction (DR) is necessary in order to develop hyperspectral sensors for real-time monitoring of large sets of samples with different characteristics. In particular, in this work, we focused on DR methods to convert the three-dimensional data array corresponding to each hyperspectral image into a one-dimensional signal (1D-DR), which retains spectral and/or spatial information. In this way, large datasets of hyperspectral images can be converted into matrices of signals, which in turn can be easily processed using suitable multivariate statistical methods. Obviously, different 1D-DR methods highlight different aspects of the hyperspectral image dataset. Therefore, in order to investigate their advantages and disadvantages, in this work, we compared three different 1D-DR methods: average spectrum (AS), single space hyperspectrogram (SSH) and common space hyperspectrogram (CSH). In particular, we have considered 370 NIR-hyperspectral images of a set of green coffee samples, and the three 1D-DR methods were tested for their effectiveness in sensor fault detection, data structure exploration and sample classification according to coffee variety and to coffee processing method. Principal component analysis and partial least squares-discriminant analysis were used to compare the three separate DR methods. Furthermore, low-level and mid-level data fusion was also employed to test the advantages of using AS, SSH and CSH altogether. Graphical Abstract Key steps in hyperspectral data dimenionality reduction.

Spectral variation is profound in remotely sensed images due to variable imaging conditions. The wide presence of such spectral variation degrades the performance of hyperspectral analysis, such as classification and spectral unmixing. In this letter, $\ell_{1}$ -based low-rank matrix approximation is proposed to alleviate spectral variation for hyperspectral image analysis. Specifically, hyperspectral image data are decomposed into a low-rank matrix and a sparse matrix, and it is assumed that intrinsic spectral features are represented by the low-rank matrix and spectral variation is accommodated by the sparse matrix. As a result, the performance of image data analysis can be improved by working on the low-rank matrix. Experiments on benchmark hyperspectral data sets demonstrate the performance of classification, and spectral unmixing can be clearly improved by the proposed approach.

Spectral unmixing is the process of decomposing the measured spectrum of a mixed pixel into a set of pure spectral signatures called endmembers and their corresponding abundances, which indicate the fractional area coverage of each endmember present in the pixel. A substantial number of spectral unmixing studies rely on a spectral mixture model which assumes that spectral mixing only occurs within the extent of a pixel. However, due to adjacency effect, the spectral measurement of the pixel may be contaminated by radiance from materials in neighboring pixels. In this paper, a linear spatial spectral mixture model that incorporates an adjacency effect in abundance estimation is proposed. We extend the classic linear mixture model by including a spatial term that expresses for each pixel the spectral contributions from its nearby pixels. An iterative optimization algorithm is developed to estimate fractional abundances of endmembers and a coefficient representing the overall intensity of the adjacency effect in the image. Our experimental results, with both synthetic and real hyperspectral images, demonstrate the effectiveness of the proposed model.

This paper proposes a blind hyperspectral reconstruction technique termed spectral compressive acquisition (SpeCA) conceived to spaceborne sensors systems which are characterized by scarce onboard computing and storage resources and by communication links with reduced bandwidth. SpeCA exploits the fact that hyperspectral vectors often belong to a low-dimensional subspace and it is blind in the sense that the subspace is learned from the measured data. SpeCA encoder is computationally very light; it just computes random projections (RPs) with the acquired spectral vectors. SpeCA decoder solves a form of blind reconstruction from RPs whose complexity, although higher than that of the encoder, is very light in the sense that it requires only the modest resources to be implemented in real time. SpeCA coding/decoding scheme achieves perfect reconstruction in noise-free hyperspectral images (HSIs) and is very competitive in noisy data. The effectiveness of the proposed methodology is illustrated in both synthetic and real scenarios.

Spectral mixture analysis (SMA) is an effective tool in recognition of unique spectral signatures of materials called endmembers and estimating their percentage of existence (abundance fractions). Most approaches designed in endmember extraction process are established by applying the spectral information of the dataset and, thus, tend to neglect the existing spatial correlation between adjacent pixels. Although several preprocessing modules have been developed by incorporating both spatial and spectral properties prior to spectral-based endmember extraction algorithms (EEs), they still encounter several challenges. Hence, in this paper, we propose an appropriate clustering and oversegmentation-based preprocessing (COPP) by greatly benefiting from the integration of spatial and spectral information. Moreover, a novel top-down oversegmentation (TDOS) algorithm is developed which can recognize small oversegments with high spatial correlation. Our scheme removes oversegments located at spatial border of cluster regions. Average spectral vectors of determined spatially homogenous oversegments are considered so that their spectral purity scores are calculated. COPP identifies spatially homogenous zones with the greatest spectral purity scores. Pixels of these regions are more likely to be adopted as endmembers by means of subsequent EEs. COPP can take advantage of degrading local spectral variability and noise power. The main contribution of this paper is the enhanced computational performance of EE as well as the precise reconstruction of the original hyperspectral scene besides its appropriate recognition of endmembers’ spectral signatures. The effectiveness of our design and its validation are appraised with the state-of-the-art strategies on a synthetic and AVIRIS real hyperspectral datasets.

Clustering for hyperspectral images (HSIs) is a very challenging task due to its inherent complexity. In this paper, we propose a novel spectral–spatial sparse subspace clustering $(\text{S}^{4}\text{C})$ algorithm for hyperspectral remote sensing images. First, by treating each kind of land-cover class as a subspace, we introduce the sparse subspace clustering (SSC) algorithm to HSIs. Then, considering the spectral and spatial properties of HSIs, the high spectral correlation and rich spatial information of the HSIs are taken into consideration in the SSC model to obtain a more accurate coefficient matrix, which is used to build the adjacent matrix. Finally, spectral clustering is applied to the adjacent matrix to obtain the final clustering result. Several experiments were conducted to illustrate the performance of the proposed $\text{S}^{4}\text{C}$ algorithm.

The hyperspectral sensors provide the images with hundreds of narrow contiguous spectral channels which provide plenty of spectral information and with this information; we can achieve high accuracy in classification problems and detection or identification of targets. However, the hyperspectral images are high in spectral resolution but low in spatial resolution. So in order to enhance the image with both spectral and spatial resolution, we support image fusion. In this paper, we have done image fusion along with preprocessing techniques such as radiometric and atmospheric corrections. The high dimensionality of the hyperspectral images may increase the computational complexity. We applied a dimensionality reduction technique that is Principal component analysis to reduce the number of bands that we consider for processing, and then performed the Image sharpening to enhance the spatial resolution. So the resulting will be high in both spectral and spatial resolution, and analysis of these image fusion techniques is also studied.

Monitoring and adjusting calibrations of various satellite imagers is often exacerbated by differences in their spectral response functions (SRFs). To help account for spectral disparities among satellite imagers, a web-based spectral band difference correction calculator has been developed to characterize the relationship between a specified pair of satellite imager channels in the hyperspectral wavelength range of 240-1750 nm. These spectral band adjustment factors (SBAFs) are derived by convolving hyperspectral data from the SCIAMACHY instrument with the SRFs of a reference and target sensor. The SBAF tool can be used for all combinations of instrument/channel pairs over predefined Earth spectra, intercalibration domains, or user-defined spatial domains. Options are available to the user whereby SBAFs can be subsetted by time, angle, and/or precipitable water content. To evaluate the relative spectral calibration of SCIAMACHY, comparisons of SBAFs derived from SCIAMACHY, Hyperion, and Global Ozone Monitoring Experiment-2 (GOME-2) were performed. Using observations over the Libya 4 desert pseudoinvariant calibration site, it is shown that SCIAMACHY-based SBAFs are within 0.1%-0.3% of SBAFs derived from Hyperion or GOME-2. This result implies that spectral calibration differences, i.e., the calibration uncertainties of SCIAMACHY relative to other potential spectral sources, have a minor impact on the SBAF compared with the influence of effective parameter-based subsetting. The SCIAMACHY instrument is most suited for calculating the SBAFs, given its high spectral resolution, broad spectral range, and nearly continuous global availability. The calibration community will find this SBAF tool useful for mitigating the SRF differences that can complicate the comparison and intercalibration of visible and near-infrared sensors.

Some spectral unmixing methods incorporate endmember variability within endmember classes. It is, however, uncertain whether these methods work well when endmember spectra do not completely describe the variability that exists within endmember classes. This paper proposes a novel spectral unmixing method, Spectral Unmixing within a multi-task Gaussian Process framework (SUGP), which is more resistant to problems caused by the use of a small number of endmember spectra. SUGP models the latent function between spectra and abundances in a training set and predicts abundances from a given pixel spectrum. SUGP is different from existing methods in that it incorporates all spectra within each endmember class to estimate abundances within a probabilistic framework. Using simulated data, SUGP was compared with existing linear unmixing methods and was found to be superior in determining the number of endmember classes within each pixel and in estimating abundances. It was also more effective in cases where a small number of spectra within endmember classes were specified and was more resistant to the effects of spectral noise. Methods were applied to the hyperspectral imagery of a mine wall and to imagery acquired over Cuprite, Nevada. Abundance maps generated by SUGP were consistent with the validated reference maps. SUGP opens up possibilities for estimating accurate abundances under conditions where endmember variability is present and where endmember spectra incompletely describe the true variability of each endmember class.