Hyperspectral Imaging For Detection Research Articles

Landmine detection in hyperspectral images based on pixel intensity

Hyperspectral imaging is a technique used to collect the same scene with different wavelengths, achieving both high spectral and spatial resolution. Hyperspectral imaging plays an important role in several scenarios involving target detection, among which landmine detection is a very challenging one. In this work, we developed a procedure based on pixel similarity measures to detect rare pixels present in a scene. The method can be combined with most of the existing detection algorithms in order to reduce the complexity and improve the performance. The developed method was tested on various types of hyperspectral images where the spectra of the landmines were simulated in different parts of the scenes with different mixing factors. The performance of the proposed method is also confirmed by tests made in real scenarios. Comparisons with state-of-the-art existing algorithms demonstrate that the method achieves excellent detection performance, with a reasonable computational complexity.

Hyperspectral two-photon excitation microscopy using visible wavelength.

We demonstrate hyperspectral imaging by visible-wavelength two-photon excitation microscopy using line illumination and slit-confocal detection. A femtosecond pulsed laser light at 530 nm was used for the simultaneous excitation of fluorescent proteins with different emission wavelengths. The use of line illumination enabled efficient detection of hyperspectral images and achieved simultaneous detection of three fluorescence spectra in the observation of living HeLa cells with an exposure time of 1 ms per line, which is equivalent to about 2 µs per pixel in point scanning, with 160 data points per spectrum. On combining linear spectral unmixing techniques, localization of fluorescent probes in the cells was achieved. A theoretical investigation of the imaging property revealed high-depth discrimination property attained through the combination of nonlinear excitation and slit detection.

Hyperspectral Anomaly Detection via Graph Dictionary-Based Low Rank Decomposition with Texture Feature Extraction

The accuracy of anomaly detection in hyperspectral images (HSIs) faces great challenges due to the high dimensionality, redundancy of data, and correlation of spectral bands. In this paper, to further improve the detection accuracy, we propose a novel anomaly detection method based on texture feature extraction and a graph dictionary-based low rank decomposition (LRD). First, instead of using traditional clustering methods for the dictionary, the proposed method employs the graph theory and designs a graph Laplacian matrix-based dictionary for LRD. The robust information of the background matrix in the LRD model is retained, and both the low rank matrix and the sparse matrix are well separated while preserving the correlation of background pixels. To further improve the detection performance, we explore and extract texture features from HSIs and integrate with the low-rank model to obtain the sparse components by decomposition. The detection results from feature maps are generated in order to suppress background components similar to anomalies in the sparse matrix and increase the strength of real anomalies. Experiments were run on one synthetic dataset and three real datasets to evaluate the performance. The results show that the performance of the proposed method yields competitive results in terms of average area under the curve (AUC) for receiver operating characteristic (ROC), i.e., 0.9845, 0.9962, 0.9699, and 0.9900 for different datasets, respectively. Compared with seven other state-of-the-art algorithms, our method yielded the highest average AUC for ROC in all datasets.

Adaptive target detection in hyperspectral imaging from two sets of training samples with different means

In this paper, we consider local detection of a target in hyperspectral imaging and we assume that the spectral signature of interest is buried in a background which follows an elliptically contoured distribution with unknown parameters. In order to infer the background parameters, two sets of training samples are available: one set, taken from pixels close to the pixel under test, shares the same mean and covariance while a second set of farther pixels shares the same covariance but has a different mean. When the whole data samples (pixel under test and training samples) follow a matrix-variate t distribution, the one-step generalized likelihood ratio test (GLRT) is derived in closed-form. It is shown that this GLRT coincides with that obtained under a Gaussian assumption and that it guarantees a constant false alarm rate. We also present a two-step GLRT where the mean and covariance of the background are estimated from the training samples only and then plugged in the GLRT based on the pixel under test only.

Generalized likelihood ratio test for modified replacement model in hyperspectral imaging detection

The replacement model, which assumes that the abundances sum up to one, is often advocated for subpixel target detection in hyperspectral imaging, and various detection schemes based on this model have been developed in the literature. However, in practical situations, this unitary constraint may be too strong due to possible attenuation of the target bidirectional reflectance distribution function, signature mismatches or impediments in the radiometric corrections. The aim of this paper is to improve the replacement-model detection algorithms by relaxing this unitary constraint. To this end, we propose to consider a modified replacement model. One step and two steps generalized likelihood ratio tests are developed for this new model and compared to standard solutions through numerical simulations. An application to real data shows the improvement offered by the proposed approach.

Sub-pixel detection in hyperspectral imaging with elliptically contoured t-distributed background

Detection of a target with known spectral signature when this target may occupy only a fraction of the pixel is an important issue in hyperspectral imaging. We recently derived the generalized likelihood ratio test (GLRT) for such sub-pixel targets, either for the so-called replacement model where the presence of a target induces a decrease of the background power, due to the sum of abundances equal to one, or for a mixed model which alleviates some of the limitations of the replacement model. In both cases, the background was assumed to be Gaussian distributed. The aim of this short communication is to extend these detectors to the broader class of elliptically contoured distributions, more precisely matrix-variate t-distributions with unknown mean and covariance matrix. We show that the generalized likelihood ratio tests in the t-distributed case coincide with their Gaussian counterparts, which confers the latter an increased generality for application. The performance as well as the robustness of these detectors are evaluated through numerical simulations.

Anomaly Detection of Hyperspectral Images Based on Linear Unmixing and Low-Rank Representation

Anomaly detection is an important branch in hyperspectral detection. There are some challenging problems mainly due to the highly mixture nature and noise corruption. In this paper, we propose a hyperspectral anomaly detection method based on Linear Unmixing and Low-Rank Representation (LU-LRR) with construction of a more robust dictionary. Firstly, the Kernel-PCA (KPCA) is used to estimate the numbers of endmembers and mine high-order correlation among spectral bands of original hyperspectral data. Secondly, data are decomposed into a product form of endmember matrix and abundance matrix based on linear spectral unmixing method. Abundance matrix usually possesses more distinctive feature to distinguish anomaly from background compared with original data. Finally, LRR is exploited to decompose abundance matrix into low-rank component and sparse component indicating background and anomalies separately, and furtherly introduce a more robust subspace basis dictionary based on dictionary learning into low-rank component. Experiments in real hyperspectral images have demonstrated that LU-LRR is more effective than traditional anomaly detection methods

In Vivo Cancer Detection in Animal Model Using Hyperspectral Image Classification with Wavelet Feature Extraction.

Hyperspectral imaging (HSI) is a promising optical imaging technique for cancer detection. However, quantitative methods need to be developed in order to utilize the rich spectral information and subtle spectral variation in such images. In this study, we explore the feasibility of using wavelet-based features from in vivo hyperspectral images for head and neck cancer detection. Hyperspectral reflectance data were collected from 12 mice bearing head and neck cancer. Catenation of 5-level wavelet decomposition outputs of hyperspectral images was used as a feature for tumor discrimination. A support vector machine (SVM) was utilized as the classifier. Seven types of mother wavelets were tested to select the one with the best performance. Classifications with raw reflectance spectra, 1-level wavelet decomposition output, and 2-level wavelet decomposition output, as well as the proposed feature were carried out for comparison. Our results show that the proposed wavelet-based feature yields better classification accuracy, and that using different type and order of mother wavelet achieves different classification results. The wavelet-based classification method provides a new approach for HSI detection of head and neck cancer in the animal model.

A Euclidean Distance based Super Resolution Method for Sub pixel target detection in Hyper Spectral Images

Most target detection algorithms suffer from the limitation that they can detect only the full pixels of the target while the target may also reside, besides the full pixel, partially in several surrounding pixels. In some cases, the target may even be embedded completely within the pixel. Both these cases are known as subpixel target detection problem. Many target detection applications, however, require detection of full pixels as well as detection of subpixel targets in the surrounding pixels which constitute a case of the mixed pixel. The problem is addressed by full pixel detection followed by spectral unmixing to determine the abundance fraction of the target. Though spectral unmixing gives the abundance fractions, it still does not give the spatial distribution/ arrangement of subpixels of the target with the surrounding pixels. The process of optimizing the spatial distribution of subpixels inside any given pixel based on the available abundance fractions is known as super resolution. This paper investigates Inverse Euclidean distance based super resolution. The algorithm performs well at different scale factors both for synthetic and real hyperspectral data which can aid the super resolution process significantly and thereby enhance the identification and recognition of target. A comparative assessment is also performed with Pixel Swap algorithm.

Game Theory-Based Hyperspectral Anomaly Detection

In this article, a novel game theory-based approach is proposed for anomaly detection in hyperspectral images (HSIs) via effectively exploring multiple spectral and spatial characteristics of anomalies. This approach comprises three main steps. First, spectral, extend morphological profiles (EMPs), and Gabor texture features are captured from an input HSI. Then, we define the anomaly detection problem as an anomaly game model, in which image regions (superpixels) of different features are modeled as players who select to be “anomaly” or “background” as their strategies. Three initial detection results are produced based on each player’s strategy in the Nash equilibrium of the anomaly game. Last, a saliency-based decision fusion technique is used to combine the complementary information in different features, so as to obtain a fused detection map. The performance of the proposed anomaly detection technique is evaluated on four real-scene HSIs. Experimental results validate that our approach can outperform some state-of-the-art anomaly detection methods.

Band selection-based collaborative representation for anomaly detection in hyperspectral images

Band selection-based collaborative representation for anomaly detection in hyperspectral images

One-Step Generalized Likelihood Ratio Test for Subpixel Target Detection in Hyperspectral Imaging

One of the main objectives of hyperspectral image processing is to detect a given target among an unknown background. The standard data to conduct such detection is a reflectance map, where the spectral signatures of each pixel's components, known as endmembers, are associated with their abundances in the pixel. Due to the low spatial resolution of most hyperspectral sensors, such a target occupies a fraction of the pixel. A widely used model in the case of subpixel targets is the replacement model. Among the vast number of possible detectors, algorithms matched to the replacement model are quite rare. One of the few examples is the finite target matched filter (MF), which is an adjustment of the well-known MF. In this article, we derive the exact generalized likelihood ratio test for this model. This new detector can be used both with a local covariance estimation window or a global one. It is shown to outperform the standard target detectors on real data, especially for small covariance estimation windows.

Near-Infrared Hyperspectral Imaging for Detection of Bacterial Fruit Blotch in Watermelon Seedlings

Near-Infrared Hyperspectral Imaging for Detection of Bacterial Fruit Blotch in Watermelon Seedlings

Portability Study of an OpenCL Algorithm for Automatic Target Detection in Hyperspectral Images

In the last decades, the problem of target detection has received considerable attention in remote sensing applications. When this problem is tackled using hyperspectral images with hundreds of bands, the use of high-performance computing (HPC) is essential. One of the most popular algorithms in the hyperspectral image analysis community for this purpose is the automatic target detection and classification algorithm (ATDCA). Previous research has already investigated the mapping of ATDCA on HPC platforms such as multicore processors, graphics processing units (GPUs), and field-programmable gate arrays (FPGAs), showing impressive speedup factors (after careful fine-tuning) that allow for its exploitation in time-critical scenarios. However, the lack of standardization resulted in most implementations being too specific to a given architecture, eliminating (or at least making extremely difficult) code reusability across different platforms. In order to address this issue, we present a portability study of an implementation of ATDCA developed using the open computing language (OpenCL). We focus on cross-platform parameters such as performance, energy consumption, and code design complexity, as compared to previously developed (hand-tuned) implementations. Our portability study analyzes different strategies to expose data parallelism as well as enable the efficient exploitation of complex memory hierarchies in heterogeneous devices. We also conduct an assessment of energy consumption and discuss metrics to analyze the quality of our code. The conducted experiments—using synthetic and real hyperspectral data sets collected by the Hyperspectral Digital Imagery Collection Experiment (HYDICE) and NASA’s Airborne Visible Infra-Red Imaging Spectrometer (AVIRIS)—demonstrate, for the first time in the literature, that portability across different HPC platforms can be achieved for real-time target detection in hyperspectral missions.

Spectral constraint adversarial autoencoders approach to feature representation in hyperspectral anomaly detection

Anomaly detection in hyperspectral images (HSIs) faces various levels of difficulty due to the high dimensionality, redundant information and deteriorated bands. To address these problems, we propose a novel unsupervised feature representation approach by incorporating a spectral constraint strategy into adversarial autoencoders (AAE) without any prior knowledge in this paper. Our approach, called SC_AAE (spectral constraint AAE), is based on the characteristics of HSIs to obtain better discrimination represented by hidden nodes. To be specific, we adopt a spectral angle distance into the loss function of AAE to enforce spectral consistency. Considering the different contribution rates of each hidden node to anomaly detection, we individually fuse the hidden nodes by an adaptive weighting method. A bi-layer architecture is then designed to suppress the variational background (BKG) while preserving features of anomalies. The experimental results demonstrate that our proposed method outperforms the state-of-the-art methods.

Spatial–spectral segmentation of hyperspectral images for subpixel target detection

Subpixel target detection in hyperspectral images (HSIs) is a problem that emerges in many important civil, military, and environmental applications. The widely used, simple in calculation, matched-filter detector is well-known to exhibit reduced performance when the target-hosting image comprises spectrally diverse, incoherent background regions. We propose to calculate and apply matched-filter subpixel target detection in image regions that exhibit both spectral and spatial coherence. Specifically, our contributions are as follows. (a) We propose the spatial–spectral expectation-maximization (SSEM) algorithm for the segmentation of a target-hosting HSI. (b) We conduct local matched filter computation and application in the SSEM-calculated image segments. The proposed method is accompanied by extensive experimental studies that corroborate its merits.

Binary-Class Collaborative Representation for Target Detection in Hyperspectral Images

Hyperspectral target detection refers to an approach that tries to locate targets in a hyperspectral image (HSI) on the condition of given targets spectrum, which plays an important role in hyperspectral remote sensing image processing. In this letter, we propose a binary-class collaborative representation-based detector. The proposed algorithm uses the concept that each background pixel can be approximately represented by its adjacent pixels within a sliding dual-window, and each target pixel can also be approximately represented by some pixels of the image; we use the given target pixels to represent it. Before estimating each background pixel, a background dictionary purification process is proposed to further improve the detector performance. The proposed algorithm was tested on three benchmark HSI data sets, and the experimental results show that the proposed algorithm demonstrates outstanding detection performances when compared with other state-of-the-art detectors.

Spectral–Spatial Discriminant Feature Learning for Hyperspectral Image Classification

Sparse representation classification (SRC) is being widely applied to target detection in hyperspectral images (HSI). However, due to the problem in HSI that high-dimensional data contain redundant information, SRC methods may fail to achieve high classification performance, even with a large number of spectral bands. Selecting a subset of predictive features in a high-dimensional space is an important and challenging problem for hyperspectral image classification. In this paper, we propose a novel discriminant feature learning (DFL) method, which combines spectral and spatial information into a hypergraph Laplacian. First, a subset of discriminative features is selected, which preserve the spectral structure of data and the inter- and intra-class constraints on labeled training samples. A feature evaluator is obtained by semi-supervised learning with the hypergraph Laplacian. Secondly, the selected features are mapped into a further lower-dimensional eigenspace through a generalized eigendecomposition of the Laplacian matrix. The finally extracted discriminative features are used in a joint sparsity-model algorithm. Experiments conducted with benchmark data sets and different experimental settings show that our proposed method increases classification accuracy and outperforms the state-of-the-art HSI classification methods.

Convolutional neural network target detection in hyperspectral imaging for maritime surveillance

This work addresses a hyperspectral imaging system for maritime surveillance using unmanned aerial vehicles. The objective was to detect the presence of vessels using purely spatial and spectral hyperspectral information. To accomplish this objective, we implemented a novel 3-D convolutional neural network approach and compared against two implementations of other state-of-the-art methods: spectral angle mapper and hyperspectral derivative anomaly detection. The hyperspectral imaging system was developed during the SUNNY project, and the methods were tested using data collected during the project final demonstration, in São Jacinto Air Force Base, Aveiro (Portugal). The obtained results show that a 3-D CNN is able to improve the recall value, depending on the class, by an interval between 27% minimum, to a maximum of over 40%, when compared to spectral angle mapper and hyperspectral derivative anomaly detection approaches. Proving that 3-D CNN deep learning techniques that combine spectral and spatial information can be used to improve the detection of targets classification accuracy in hyperspectral imaging unmanned aerial vehicles maritime surveillance applications.

A new adaptive algorithm for target detection in hyperspectral images

In this paper, a novel adaptive algorithm for target detection in hyperspectral images (HSIs) is proposed. In a general classification, the proposed method belongs to the category of those methods which are not based on the statistical moments of the observed HSI (e.g. correlation or covariance matrix). The main processing burden of the proposed method is over a known set of spectral signatures. Assuming a linear spectral mixing model, the proposed method takes a set of spectral signatures which one of them relates to the target material and the others relate to the background materials. Based on an adaptive approach, the normalized least mean square (NLMS) adaptive algorithm is engaged to estimate a weight vector which is almost orthogonal to the background materials spectral signature whereas it makes an absolutely non-orthogonal pair with the target material spectral signature. The estimated weight vector is multiplied by the observed HSI to make the final decision. One synthetic and two real hyperspectral images are considered to evaluate the performance of the proposed method. The evaluation results show that the proposed method outperforms its counterparts.