Collaborative dictionary learning with structured incoherence for target detection in hyperspectral imagery

In this paper we present a new methodology for automated target detection and identification in hyperspectral imagery. The standard paradigm for target detection in hyperspectral imagery is to run a detection algorithm, typically statistical in nature, and visually inspect each high-scoring pixel to decide whether it is a true detection or a false alarm. Detection filters have constant false alarm rates (CFARs) approaching 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-5</sup> , but these can still result in a large number of false alarms given multiple images and a large number of target materials. Here we introduce a new methodology for target detection and identification in hyperspectral imagery in conjunction with a data-driven target detection algorithm that shows promise for hard targets. The result is a greatly reduced false alarm rate and a practical methodology for aiding an analyst in quantitatively evaluating detected pixels.

Sparse representation based classification (SRC) and collaborative representation based classification (CRC) are two well-known methods in representation-based classification for face recognition. SRC emphasizes on the role of sparsity in coding, while CRC obtains an enhanced sparse representation with collaborative contributions from all classes. Previous studies show that sparsity of representation coefficients of SRC and CRC is significant for robust classification that may amplify the nonlocal contribution from other classes. However, the parameters are closely associated with the sparsity and it is very difficult to solve the optimal solutions. This paper proposed a novel representation-based image classification method that integrates SRC with CRC by a multiplication operation on the solutions to obtain enhanced sparse representation. We conducted experiments on several benchmark face databases and the experimental results showed that our algorithm produces higher accuracy on classification than both SRC and CRC.

Kernel-based regularized-angle spectral matching for target detection in hyperspectral imagery

In this paper, we propose a collaborative representation-based binary hypothesis model with multi-features learning (CRTDBH-MTL) for target detection in hyperspectral imagery. The proposed method contained the following aspects. First, two complementary features extracted by different algorithms are implemented for describing hyperspectral imageries. Next, we apply these features into the unified collaborative representation-based binary hypothesis model (CRTDBH) to acquire a collaborative vector (CV) for each feature. Once the CV is obtained, the sample can be sparsely represented by the training samples from the background-only dictionary under the null hypothesis and the training samples from the target and background dictionaries under the alternative hypothesis. Finally, spatial correlation and spectral similarity of adjacent neighboring pixels are exploited to improve the detection performance. The experimental results suggest that the proposed algorithm shows an outstanding detection performance.

In recent years, representation-based methods have attracted more attention in the hyperspectral image (HSI) classification. Among them, sparse representation-based classifier (SRC) and collaborative representation-based classifier (CRC) are the two representative methods. However, SRC only focuses on sparsity but ignores the data correlation information. While CRC encourages grouping correlated variables together but lacks the ability of variable selection. As a result, SRC and CRC are incapable of producing satisfied performance. To address these issues, in this work, a correlation adaptive representation (CAR) is proposed, enabling a CAR-based classifier (CARC). Specifically, the proposed CARC is able to explore sparsity and data correlation information jointly, generating a novel representation model that is adaptive to the structure of the dictionary. To further exploit the correlation between the test samples and the training samples effectively, a distance-weighted Tikhonov regularization is integrated into the proposed CARC. Furthermore, to handle the small training sample problem in the HSI classification, a multi-feature correlation adaptive representation-based classifier (MFCARC) and MFCARC with Tikhonov regularization (MFCART) are presented to improve the classification performance by exploring the complementary information across multiple features. The experimental results show the superiority of the proposed methods over state-of-the-art algorithms.

During the last decade, sparse representations have been successfully applied to design high-performing classification algorithms such as the classical sparse representation based classification (SRC) algorithm. More recently, collaborative representation based classification (CRC) has emerged as a very powerful approach, especially for face recognition. CRC takes advantage of SRC through the notion of collaborative representation, relying on the observation that the collaborative property is more crucial for classification than the l1-norm sparsity constraint on coding coefficients used in SRC. This paper follows the same general philosophy of CRC and its main novelty is the application of a virtual collaborative projection (VCP) routine designed to train images of every class against the other classes to improve fidelity before the projection of the query image. We combine this routine with a method of local feature extraction based on high-order statistical moments to further improve the representation. We demonstrate using extensive experiments of face recognition and classification that our approach performs very competitively with respect to state-of-the-art classification methods. For instance, using the AR face dataset, our method reaches 100% of accuracy for dimensionality 300.

Orthogonal subspace projection (OSP) has been used in hyperspectral image processing for automatic target detection and image classification. Existing OSP based approaches for target detection require a priori knowledge of all undesired signatures present in the input scene. In this paper, we proposed a new technique for target detection which does not require a priori knowledge of the non-target signatures present in the input scene. The length of any pixel vector containing the target reduces significantly when it is projected in a direction orthogonal to the target signature. Thus the ratio between the original pixel vector to the projected pixel vector yields a high value for the pixels containing the target. Therefore, an OSP based parameter along with noise adjusted principal component analysis (NAPCA) was introduced in this paper for target detection in hyperspectral images. For noisy images, NAPCA is used as a preprocessing step to reduce the effects of noise as well as to reduce the spectral dimension thereby yielding better target detection capability while enhancing the computational efficiency. For noise-free input scenes or when very small amount of noise is present in the input scene, principal component analysis (PCA) may be used instead of NAPCA. The OSP based technique requires that the number of spectrally distinct signatures present in the input scene must be less than the number of spectral bands. The proposed algorithm yields very good results even when this criterion is not satisfied.

Metric learning-based methods, which yield great performance and show considerable potential to improve the performance of hyperspectral image processing, aim to calculate the Mahalanobis distance metric matrix. In this article, we proposed a symmetric information-theoretic metric learning (SITML) method for hyperspectral target detection. The SITML algorithm is designed based on the classical information-theoretic metric learning (ITML) and, minimizes the differential Kullback–Leibler (KL) divergence. To enhance both of the detection performance and the generalization ability, we build metric spaces from the neighborhood of training samples to preserve the local discriminative information. Then, we conduct local pairwise constraints to maximize the Jeffery divergence (also named the symmetric KL divergence) of two multivariate Gaussian distributions to solve the problem of an asymmetric KL divergence. Finally, we use a closed-form solution to solve the optimization problem. Intensive experiments on three hyperspectral datasets indicate that SITML outperforms the classical ITML algorithm and other state-of-the-art target detection methods.

Target detection in hyperspectral imagery is a complex process due to many factors. Exploiting the hyperspectral image for analysis is very challenging due to large information and low spatial resolution. However, hyperspectral target detection has numerous applications. Hence, it is important to pursue research in target detection. In this paper, a literature survey of target detection algorithms for hyperspectral imagery is presented along with scope for future research. An overview of theory behind the hyperspectral imaging is detailed. Also, various challenges involved in exploring the hyperspectral data are discussed.

This paper introduces Cointegration Theory to address the problem of adaptive target detection in hyperspectral imagery. Cointegration Theory aims at mining a long-term equilibrium relationship, which refers to the condition that an appropriate linear combination of several non-stationary series can be stationary as long as they have similar or related drift. Hyperspectral response sequences, which are highly non-stationary, have similar patterns among the same materials. To be treated as a time series, each given hyperspectral curve is matched with the reference spectrum via the Johansen Cointegration Test. The statistic of the test is then used for target detection. Experimental results indicate that our proposed method is effective and has a strong capacity to identify interesting objects from their background.

Unstructured detectors such as KGLRT, ACE and AMF are widely applied for target detection in hyperspectral imagery (HSI). However, conventional global and local approaches construct background model without considering the contamination caused by anomalies and suspected targets. This paper proposes a local ACE algorithm based on the minimum covariance determinant (MCD) estimator. In the proposed algo-rithm, a spectral angle based clustering method is applied to the whitened hyperspectral data to form several disjoint clusters over the whole image. Then for each cluster, the robust estimations of its background statistics are obtained using the MCD estimator. Finally, the ACE detector is applied to each pixel utilizing the robust background statistics of the cluster. With experimental results on two different real datasets, the superiority of the proposed algorithm is demonstrated.

Subpixel target detection in hyperspectral imagery is challenging since subpixel targets are smaller in size than the resolution of a single pixel and accurate pixel-level labels on subpixel targets are often unavailable. In particular, this article addresses the problem of learning a prime prototype target signature from imprecisely labeled highly mixed hyperspectral data. Two algorithms, multiple-instance subpixel adaptive cosine estimator (MI-SPACE) and multiple-instance subpixel spectral matched filter (MI-SPSMF), based on multiple-instance learning framework are presented. The proposed methods aim to learn a discriminative prime target signature by maximizing the posterior detection statistics of subpixel hyperspectral targets for the correspondingly proposed subpixel adaptive cosine estimator (SPACE) and subpixel spectral matched filter (SPSMF) detectors, which are also developed in this article. Experimental results demonstrate the effectiveness of the proposed methods on both simulated and real-field hyperspectral subpixel target detection tasks.

Sparse representation based on stacked kernel for target detection in hyperspectral imagery

Mapping cotton cultivated area combining remote sensing with a fused representation-based classification algorithm

Collaborative representation based classification (CRC) is an efficient classifier in image classification. By using regularization, the collaborative representation based classifier holds competitive performances compared with the sparse representation based classifier using less computational time. However, each of the elements calculated from the training samples are utilized for representation without selection, which can lead to poor performances in some classification tasks. To resolve this issue, in this paper, we propose a novel collaborative representation by directly using non-negative representations to represent a test sample collaboratively, termed Non-negative Collaborative Representation-based Classifier (NCRC). To collect all non-negative collaborative representations, we introduce a Rectified Linear Unit (ReLU) function to perform filtering on the coefficients obtained by minimization according to CRC’s objective function. Next, we represent the test sample by using a linear combination of these representations. Lastly, the nearest subspace classifier is used to perform classification on the test samples. The experiments performed on four different databases including face and palmprint showed the promising results of the proposed method. Accuracy comparisons with other state-of-art sparse representation-based classifiers demonstrated the effectiveness of NCRC at image classification. In addition, the proposed NCRC consumes less computational time, further illustrating the efficiency of NCRC.