Over the last two decades, anomaly detection (AD) has been known to play a critical role in hyperspectral image analysis, which provides a new way to distinguish the targets from the background without prior knowledge. Recently, the representation-based methods were proposed and soon became a significant type of methods on hyperspectral AD. In this paper, a novel AD algorithm based on convolutional neural network (CNN) and low-rank representation (LRR) is proposed. First, a CNN model is built and trained on hyperspectral image (HSI) datasets to accurately obtain the resulting abundance maps. Compared with the raw dataset, abundance maps contain more distinctive features to identify anomalies from the background. Second, a dictionary is constructed by the density-based spatial clustering of applications with noise (DBSCAN) algorithm to stably represent the background component. Third, a matrix decomposition method based on LRR is adopted. In this way, a coefficient matrix corresponding to the constructed dictionary is obtained, which is low rank. At the same time, a residual matrix can be obtained as well, which is sparse. Finally, anomalies can be extracted from the residual matrix. The experimental results show that the proposed method achieves a superior performance compared to some of the state-of-the-art methods in the field of hyperspectral AD.

Visual perception system is the key component of safety driving assistance and unmanned driving systems, and the perceptive performance of which will directly affect the running safety. There are various perception algorithms for the visual system, especially in recent years, deep learning algorithms are adopted with increasing popularity among them. However, the majority of existing works mainly focuses on daytime scenes with favorable illumination and weather conditions, and relies on visible light with intensity imaging equipment. This paper is based on polarization features acquired by deep learning methods, attempting to address the problems about vehicle-road environment perception in Low-Visibility Condition. Polarization is one of the essential properties of matter, here we propose a new method by studying and analyzing the transmission characteristics of the target’s polarization information. In the proposed method, the polarization features of vehicle-road environment are first acquired by a kind of three-channel polarization imaging device, then, fused together with an intensity image, and finally followed by a semantic segmentation operation by deep networks. The experimental results clearly demonstrate that the effect of perception based on polarization features via deep learning are greatly improved compared with intensity imaging, especially in low illumination and severe weather conditions. It is of great value for environmental robustness of visual perception system and further traffic safety.

Various existing anomaly detection (AD) technologies focus on the background feature extraction and suppression, which serves as a crucial step to extrude anomalies from the hyperspectral imagery (HSI). In this article, motivated by the advantages of the joint sparse representation (JSR) model for adaptive background base selection, a robust background feature extraction method through homogeneous region-based JSR is proposed and used for AD. By segmenting the scene from the spatial domain through an eight-connected region division operation based on the clustering result, a series of nonoverlapping homogeneous regions each sharing a common sparsity pattern are obtained. After discarding small regions, JSR is performed on each region with the dictionary constituted by the overall spectral items in the corresponding cluster. By calculating the usage frequency of dictionary atoms, the most representative background bases describing each background cluster are adaptively selected and then combined into the global background bases. In addition, considering the interference of noise on detection accuracy, an energy deviation-based noise estimation strategy is presented by analyzing the residual obtained from JSR. Finally, the anomaly response of each pixel is measured by comparing its projection energy obtained from the background orthogonal subspace projection with the noise energy in its corresponding region. The proposed method overcomes the shortcomings of traditional neighborhood-based JSR in the common sparsity pattern and anomaly proportion. The spatial characteristics of HSI are fully explored. Furthermore, the interference of noise on detection accuracy is eliminated. Experiments on four HSI data sets demonstrate the superiority of the proposed method.

Although the low-rank and sparse matrix decomposition (LRaSMD)-based anomaly detectors can effectively extract the low-rank structure as the background component and the sparse structure as the anomaly component for anomaly detection (AD) while simultaneously considering the additive noise, the background interferences in the sparse component remain a serious problem that will increase the false alarm rate and influence the detection of real anomalies. To alleviate this issue, a novel LRaSMD with orthogonal subspace projection (OSP)-based background suppression and adaptive weighting for hyperspectral AD is proposed in this letter. Based on the fact that the background interferences in the sparse component are mainly some sparse objects with slight spectral differences from the main background, the OSP is employed to project the sparse component into the background orthogonal subspace that is estimated from the low-rank component to suppress the background interferences and highlight the anomalies. Furthermore, the low-rank component provides an effective estimation of the background statistics, which can be used to adaptively weigh the detection results. Experiments on both synthetic and real hyperspectral data sets demonstrate the effectiveness of the proposed algorithm.

Anomaly detection (AD), which aims to distinguish targets with significant spectral differences from the background, has become an important topic in hyperspectral imagery (HSI) processing. In this paper, a novel anomaly detection algorithm via dictionary construction-based low-rank representation (LRR) and adaptive weighting is proposed. This algorithm has three main advantages. First, based on the consistency with AD problem, the LRR is employed to mine the lowest-rank representation of hyperspectral data by imposing a low-rank constraint on the representation coefficients. Sparse component contains most of the anomaly information and can be used for anomaly detection. Second, to better separate the sparse anomalies from the background component, a background dictionary construction strategy based on the usage frequency of the dictionary atoms for HSI reconstruction is proposed. The constructed dictionary excludes possible anomalies and contains all background categories, thus spanning a more reasonable background space. Finally, to further enhance the response difference between the background pixels and anomalies, the response output obtained by LRR is multiplied by an adaptive weighting matrix. Therefore, the anomaly pixels are more easily distinguished from the background. Experiments on synthetic and real-world hyperspectral datasets demonstrate the superiority of our proposed method over other AD detectors.

Due to the limited resolution of hyperspectral sensors, anomalous targets expressed with subpixels are often mixed with nonhomogeneous backgrounds. This fact makes anomalies difficult to distinguish from the surrounding background. From this perspective, a novel hyperspectral anomaly detection (AD) algorithm based on endmember extraction and low-rank representation (LRR) is proposed. For the characteristics of pixels in hyperspectral images (HSIs), the proposed algorithm employs an endmember extraction technology to yield an abundance matrix for AD, thereby gathering more feature information compared with the direct use of a raw image. In addition, a dictionary construction strategy based on Tucker decomposition, and the k-means++ clustering method is proposed to make the dictionary more stable and discriminative. An LRR method based on the dictionary is applied to obtain a sparse residual matrix. Finally, anomalies can be determined by the response of the residual matrix. Experiments on three hyperspectral data sets validate the performance of the proposed algorithm.