Qiong Ran

Weighted-RXD and Linear Filter-Based RXD: Improving Background Statistics Estimation for Anomaly Detection in Hyperspectral Imagery

Anomaly detection is an active topic in hyperspectral imaging, with many practical applications. Reed-Xiaoli detector (RXD), a widely used method for anomaly detection, uses the covariance matrix and mean vector to represent background signals, assuming that the background information adjusts to a multivariate normal distribution. However, in general, real images present very complex backgrounds. As a result, in many situations, the background information cannot be properly modeled. An important reason is that that background samples often contain also anomalous pixels and noise, which lead to a high false alarm rate. Therefore, the characterization of the background is essential for successful anomaly detection. In this paper, we develop two novel approaches: weighted-RXD (W-RXD) and linear filter-based RXD (LF-RXD) aimed at improving background in RXD-based anomaly detection. By reducing the weight of the anomalous pixels or noise signals and increasing the weight of the background samples, W-RXD can provide better estimations of the background information. In turn, LF-RXD uses the probability of each pixel as background to filter wrong anomalous or noisy instances. Our experimental results, intended to analyze the performance of the newly developed anomaly detectors, indicate that the proposed approaches achieve good performance when compared with other classic approaches for anomaly detection in the literature.

Hyperspectral Image Classification Using Weighted Joint Collaborative Representation

Recently, representation-based classifiers have gained increasing interest in hyperspectral image (HSI) classification. In this letter, based on our previously developed joint collaborative representation (JCR) classifier, an improved version, which is called weighted JCR (WJCR) classifier, is proposed. JCR adopts the same weights when extracting spatial and spectral features from surrounding pixels. Differing from JCR, WJCR attempts to utilize more appropriate weights by considering the similarity between the center pixel and its surroundings. Experimental results using two real HSIs demon strate that the proposed WJCR outperforms the original JCR and some other traditional classifiers, such as the support vector machine (SVM), the SVM with a composite kernel, and simultaneous orthogonal matching pursuit.

PSO-EM: A Hyperspectral Unmixing Algorithm Based On Normal Compositional Model

A new hyperspectral unmixing algorithm is proposed based on the normal compositional model (NCM) to estimate the endmembers and abundance parameters jointly in this paper. The NCM considers the hyperspectral imaging as a stochastic process and interprets each pixel value as a random vector, which is linearly mixed by the endmembers. More precisely, these endmembers are also treated as random variables as opposed to deterministic values in order to capture spectral variability that is not well described by the linear mixing model (LMM). However, the higher complexity of such an unmixing model leads to more difficulty in parameter estimation. A particle swarm optimization-expectation maximization (PSO-EM) algorithm, a “winner-take-all” version of the EM, is proposed to solve the parameter estimation problem, which employs a partial E step. The main contribution of the proposed PSO-EM is making optimum use of particle swarm optimization method (PSO) in the partial E step, which solves the difficulty of the integrals in the NCM model. The performance of the proposed methodology is evaluated through synthetic and real data experiments. Our obtained results demonstrate the superior performance of PSO-EM compared to other NCM-based as well as LMM-based methods.

Residual-scaled local standard deviations method for estimating noise in hyperspectral images

A new method for estimating noise in hyperspectral images is described in this paper. It is based on the strong between-band correlation of hyperspectral images and the concept of local standard deviations of small imaging blocks. The new method can be used to automatically estimate noise for both radiance and reflectance images. Unlike other methods discussed in this paper, the new method is more reliable for estimating noise in hyperspectral images with diverse land cover types. We successfully applied the new method in estimating noise for Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data.

Hyperspectral image classification with improved local-region filters

Abstract Two improved local-region filters, adaptive weighted filter (AWF) and collaborative representation filter (CoRF), are proposed for feature extraction and classification in hyperspectral imagery. The local-region filters generate spatial-spectral features of a hyperspectral pixel by incorporating its surrounding pixels. The work of this paper is an extension of our previously introduced local average filter (LAF). Unlike LAF, which gives the surrounding pixels the same weight, AWF and CoRF explore the internal similarity in the local region with an adaptive weight. More specifically, AWF is set up considering the spatial distance to the central pixel, and CoRF is constructed with spectral similarities adopting the idea of collaborative representation. The two improved local-region filters adaptively extract spectral-spatial features from neighboring pixels and are proven to be effective in many aspects, such as edge information preservation and classification performance, with experiments on two real hyperspectral datasets.

Hyperspectral image classification for mapping agricultural tillage practices

Abstract. An efficient classification framework for mapping agricultural tillage practice using hyperspectral remote sensing imagery is proposed, which has the potential to be implemented practically to provide rapid, accurate, and objective surveying data for precision agricultural management and appraisal from large-scale remote sensing images. It includes a local region filter [i.e., Gaussian low-pass filter (GLF)] to extract spatial-spectral features, a dimensionality reduction process [i.e., local fisher’s discriminate analysis (LFDA)], and the traditional k-nearest neighbor (KNN) classifier, and is denoted as GLF-LFDA-KNN. Compared to our previously used local average filter and adaptive weighted filter, the GLF also considers spatial features in a small neighborhood, but it emphasizes the central pixel itself and is data-independent; therefore, it can achieve the balance between classification accuracy and computational complexity. The KNN classifier has a lower computational complexity compared to the traditional support vector machine (SVM). After classification separability is enhanced by the GLF and LFDA, the less powerful KNN can outperform SVM and the overall computational cost remains lower. The proposed framework can also outperform the SVM with composite kernel (SVM-CK) that uses spatial-spectral features.

Multisource Remote Sensing Data Classification Based on Convolutional Neural Network

As a list of remotely sensed data sources is available, how to efficiently exploit useful information from multisource data for better Earth observation becomes an interesting but challenging problem. In this paper, the classification fusion of hyperspectral imagery (HSI) and data from other multiple sensors, such as light detection and ranging (LiDAR) data, is investigated with the state-of-the-art deep learning, named the two-branch convolution neural network (CNN). More specific, a two-tunnel CNN framework is first developed to extract spectral-spatial features from HSI; besides, the CNN with cascade block is designed for feature extraction from LiDAR or high-resolution visual image. In the feature fusion stage, the spatial and spectral features of HSI are first integrated in a dual-tunnel branch, and then combined with other data features extracted from a cascade network. Experimental results based on several multisource data demonstrate the proposed two-branch CNN that can achieve more excellent classification performance than some existing methods.

Representation-based classifications with Markov random field model for hyperspectral urban data

Abstract Recently, representation-based classifications have gained increasing interest in hyperspectral imagery, such as the newly proposed sparse-representation classification and nearest-regularized subspace (NRS). These classifiers provide excellent performance that is comparable to or even better than the classic support vector machine. However, all these representation-based methods were originally designed to be pixel-wise classifiers which only consider the spectral signature while ignoring the spatial-contextual information. A Markov random field (MRF), providing a basis for modeling contextual constraints, has currently been successfully applied for hyperspectral image analysis. We mainly investigate the benefits of combining these representation-based classifications with an MRF model in order to acquire better classification results. Two real hyperspectral images are used to validate the proposed classification scheme. Experimental results demonstrated that the proposed method significantly outperforms other state-of-the-art approaches. For example, NRS-MRF performed with an accuracy of 94.92% for the Reflective Optics System Imaging Spectrometer data with 60 training samples per class, while the original NRS obtained an accuracy of 81.95%, an improvement of approximately 13%.

Parallel computation for blood cell classification in medical hyperspectral imagery

With the advantage of fine spectral resolution, hyperspectral imagery provides great potential for cell classification. This paper provides a promising classification system including the following three stages: (1) band selection for a subset of spectral bands with distinctive and informative features, (2) spectral-spatial feature extraction, such as local binary patterns (LBP), and (3) followed by an effective classifier. Moreover, these three steps are further implemented on graphics processing units (GPU) respectively, which makes the system real-time and more practical. The GPU parallel implementation is compared with the serial implementation on central processing units (CPU). Experimental results based on real medical hyperspectral data demonstrate that the proposed system is able to offer high accuracy and fast speed, which are appealing for cell classification in medical hyperspectral imagery.

Improving hyperspectral image classification using smoothing filter via sparse gradient minimization

In hyperspectral imagery, there exist homogeneous regions where neighboring pixels tend to belong to the same class with high probability. However, even though neighboring pixels are from the same material, their spectral characteristics may be different due to various factors, such as internal instrument noise or atmospheric scattering, which results in misclassification. In this work, the proposed framework employs a smoothing filter based on sparse gradient minimization, which is expected to eliminate the inherent variations within a small neighborhood. Experimental results for two hyperspectral image datasets demonstrate that the proposed algorithm significantly improve classification accuracy.