Hyperspectral anomaly detection based on constrained eigenvalue–eigenvector model

Anomaly detection in a large area using hyperspectral imaging is an important application in real-time remote sensing. Anomaly detectors based on subspace models are suitable for such an anomaly and usually assume the main background subspace and its dimensions are known. These detectors can detect the anomaly for a range of values of the dimension of the subspace. The objective of this paper is to develop an anomaly detector that extends this range of values by assuming main background subspace with an unknown user-specified dimension and by imposing covariance of error to be a diagonal matrix. A pixel from the image is modeled as the sum of a linear combination of the unknown main background subspace and an unknown error. By having more unknown quantities, there are more degrees of freedom to find a better way to fit data to the model. By having a diagonal matrix for the covariance of the error, the error components become uncorrelated. The coefficients of the linear combination are unknown, but are solved by using a maximum likelihood estimation. Experimental results using real hyperspectral images show that the anomaly detector can detect the anomaly for a significantly larger range of values for the dimension of the subspace than conventional anomaly detectors.     

Variable factorization model based on numerical optimization for hyperspectral anomaly detection

The objective of this article is to develop an anomaly detector as an analytical expression for detecting anomalous objects in remote sensing using hyperspectral imaging. Conventional anomaly detectors based on the subspace model have a parameter which is the dimension of the clutter subspace. The range of possible values for this parameter is typically large, resulting in a large number of images of detector output to be analyzed. An anomaly detector with a different parameter is proposed. The pixel of known random variables from a data cube is modeled as a linear transformation of a set of unknown random variables from the clutter subspace plus an error of unknown random variables in which the transformation matrix of constants is also unknown. The dimension of the clutter subspace for each spectral component of the pixel can vary, hence some elements in the transformation matrix are constrained to be zeros. The anomaly detector is the Mahalanobis distance of the resulting residual. The experimental results which are obtained by implementing the anomaly detector as a global anomaly detector in unsupervised mode with background statistics computed from hyperspectral data cubes with wavelengths in the visible and near-infrared range show that the parameter in the anomaly detector has a significantly reduced number of possible values in comparison with conventional anomaly detectors.     

Maximized subspace model for hyperspectral anomaly detection

An important application in remote sensing using hyperspectral imaging system is the detection of anomalies in a large background in real-time. A basic anomaly detector for hyperspectral imagery that performs reasonaly well is the RX detector. In practice, the subspace RX (SSRX) detector which deletes the clutter subspace has been known to perform better than the RX detector. In this paper an anomaly detector that can do better than the SSRX detector without having to delete the clutter subspace is developed. The anomaly detector partials out the effect of the clutter subspace by predicting the background using a linear combination of the clutter subspace. The Mahalanobis distance of the resulting residual is defined as the anomaly detector. The coefficients of the linear combination are chosen to maximize a criterion based on squared correlation. The experimental results are obtained by implementing the anomaly detector as a global anomaly detector in unsupervised mode with background statistics computed from hyperspectral data cubes with wavelengths in the visible and near-infrared range. The results show that the anomaly detector has a better performance than the SSRX detector. In conclusion, the anomaly detector that is based on partialling out can achieve better performance than the conventional anomaly detectors.     

改进协同表示的高光谱图像异常检测算法

针对协同表示的高光谱图像异常检测算法中双窗口中心为异常像元同时背景字典存在同种异常像元的情况,中心像元的输出较小难以与背景区分的问题,提出一种改进协同表示的高光谱图像异常检测算法。为了减小背景字典中异常像元的权重,使用背景字典原子与均值的距离调整原子的权重,从而增大上述情况下中心像元的输出。实验结果表明,提出的算法在不同双窗口下都取得了较好的检测效果,验证了算法的有效性。     

Random selection-based adaptive saliency-weighted RXD anomaly detection for hyperspectral imagery

With recent advances in hyperspectral imaging sensors, subtle and concealed targets that cannot be detected by multispectral imagery can be identified. The most widely used anomaly detection method is based on the Reed–Xiaoli (RX) algorithm. This unsupervised technique is preferable to supervised methods because it requires no a priori information for target detection. However, two major problems limit the performance of the RX detector (RXD). First, the background covariance matrix cannot be properly modelled because the complex background contains anomalous pixels and the images contain noise. Second, most RX-like methods use spectral information provided by data samples but ignore the spatial information of local pixels. Based on this observation, this article extends the concept of the weighted RX to develop a new approach called an adaptive saliency-weighted RXD (ASW-RXD) approach that integrates spectral and spatial image information into an RXD to improve anomaly detection performance at the pixel level. We recast the background covariance matrix and the mean vector of the RX function by multiplying them by a joint weight that in fuses spectral and local spatial information into each pixel. To better estimate the purity of the background, pixels are randomly selected from the image to represent background statistics. Experiments on two hyperspectral images showed that the proposed random selection-based ASW RXD (RSASW-RXD) approach can detect anomalies of various sizes, ranging from a few pixels to the sub-pixel level. It also yielded good performance compared with other benchmark methods.     

微型航天探测器目标识别算法研究

微型航天目标探测器主要是对空间目标探测识别,然而在微型航天探测器在空间对目标物成像时目标物、背景和探测器之间存在相对运动,目前还没有好的运动背景中运动目标识别的算法.因此,针对微型航天探测器在星空背景下对目标物识别的要求,提出了把图像配准与双差分法相结合目标识别的算法.该算法对连续采集的三帧图像中的前两帧和后两帧分别进行配准,差分,二值化,然后将两副二值化图像逻辑乘,识别出目标.计算机仿真的结果表明该算法计算量比较小,对目标的识别效果好.因而,基于图像配准的双差分法是一个相对较好适用于微型航天探测器对空间目标探测识别的算法.     

结合孪生网络和像素配对的高光谱图像异常检测

目的 高光谱遥感中,通常利用像素的光谱特征来区分背景地物和异常目标,即通过二者之间的光谱差异来寻找图像中的异常像元。但传统的异常检测算法并未有效挖掘光谱的深层特征,高光谱图像中丰富的光谱信息没有被充分利用。针对这一问题,本文提出结合孪生神经网络和像素配对策略的高光谱图像异常检测方法,利用深度学习技术提取高光谱图像的深层非线性特征,提高异常检测精度。方法 采用像素配对的思想构建训练样本,与原始数据集相比,配对得到的新数据集数量呈指数增长,从而满足深度网络对数据集数量的需求。搭建含有特征提取模块和特征处理模块的孪生网络模型,其中,特征处理模块中的卷积层可以专注于提取像素对之间的差异特征,随后利用新的训练像素对数据集进行训练,并将训练好的分类模型固定参数,迁移至检测过程。用滑动双窗口策略对测试集进行配对处理,将测试像素对数据集送入网络模型,得到每个像素相较于周围背景像素的差异性分数,从而识别测试场景中的异常地物。结果 在异常检测的实验结果中,本文提出的孪生网络模型在San Diego数据集的两幅场景和ABU-Airport数据集的一幅场景上,得到的AUC （area under the curve）值分别为0.993 51、0.981 21和0.984 38,在3个测试集上的表现较传统方法和基于卷积神经网络的异常检测算法具有明显优势。结论 本文方法可以提取输入像素对的深层光谱特征,并根据其特征的差异性,让网络学习到二者的区分度,从而更好地赋予待测像素相对于周围背景的异常分数。本文方法相对于卷积神经网络的异常检测方法可以有效地降低虚警,与传统方法相比能够更加明显地突出异常目标,提高了检测率,同时也具有较强的鲁棒性。     

空域协同自编码器的高光谱异常检测

目的 自编码器作为一种无监督的特征提取算法,可以在无标签的条件下学习到样本的高阶、稠密特征。然而当训练集含噪声或异常时,会迫使自编码器学习这些异常样本的特征,导致性能下降。同时,自编码器应用于高光谱图像处理时,往往会忽略掉空域信息,进一步限制了自编码器的探测性能。针对上述问题,本文提出一种基于空域协同自编码器的高光谱异常检测算法。方法 利用块图模型优良的背景抑制能力从空域角度筛选用于自编码器训练的背景样本集。自编码器采用经预筛选的训练样本集进行网络参数更新,在提升对背景样本表达能力的同时避免异常样本对探测性能的影响。为进一步将空域信息融入探测结果,利用块图模型得到的异常响应构建权重,起到突出目标并抑制背景的作用。结果 实验在3组不同尺寸的高光谱数据集上与5种代表性的高光谱异常检测算法进行比较。本文方法在3组数据集上的AUC(area under the curve)值分别为0.990 4、0.988 8和0.997 0,均高于其他算法。同时,对比了不同的训练集选择策略,与随机选取和使用全部样本进行对比。结果表明,本文基于空域响应的样本筛选方法相较对比方法具有较明显的优势。结论 提出的基...     

粗定位和协同表示的高光谱图像异常检测

目的 由于在军事和民用应用中的重要作用,高光谱遥感影像异常检测在过去的20~30年里一直都是备受关注的研究热点。然而,考虑到异常点往往藏匿于大量的背景像元之中,且只占据很少的数量,给精确检测带来了不小的挑战。针对此问题,基于异常点往往表现在高频的细节区域这一前提,本文提出了一种基于异常点粗定位和协同表示的高光谱遥感影像异常检测算法。方法 对输入的原始高光谱遥感影像进行空间维的降质操作;通过衡量降质后影像与原始影像在空间维的差异,粗略定位可能的异常点位置;将粗定位的异常点位置用于指导像元间的协同表示以重构像元;通过衡量重构像元与原始像元的差异,从而进一步优化异常检测结果。结果 在4个数据集上与6种方法进行了实验对比。对于San Diego数据集,次优算法和本文算法分别取得的AUC （area under curve）值为0.978 6和0.994 0;对于HYDICE （hyperspectral digital image collection equipment）数据集,次优算法和本文算法的AUC值为0.993 6和0.998 5;对于Honghu数据集,次优算法和本文方法的AUC值分别为0.999 2和0.999 3;对Grand Isle数据集而言,尽管本文方法以0.001的差距略低于性能第1的算法,但从目视结果图中可见,本文方法所产生的虚警目标远少于性能第1的算法。结论 本文所提出的粗定位和协同表示的高光谱异常检测算法,综合考虑了高光谱遥感影像的谱间特性,同时还利用了其空间特性以及空间信息的先验分布,从而获得异常检测结果的提升。     

遥感异常目标的仿生非线性滤波检测

目的 为了解决复杂背景干扰下基于线性滤波异常检测算法无法有效区分复杂背景特征与异常目标特征,导致检测结果虚警率偏高的问题,提出一种面向复杂背景的遥感异常小目标仿生非线性滤波检测算法。方法 受生物视觉系统利用不同属性信息挖掘高维特征机理的启发,该算法通过相关型非线性滤波器综合多波段光谱数据提取高维光谱变化特征作为异常目标检测检测依据,弥补线性滤波抗噪性能差,难于区分复杂背景特征与目标特征的缺点。结果 仿真实验结果验证该算法在仿真数据及真实遥感数据的异常检测效果上有较大改善,在实现快速异常检测的同时提高了检测命中率。结论 本文方法不涉及背景建模,计算复杂度低,具有较好的实时性与普适性。特别是对复杂背景下的小尺寸异常目标具有较好的检测效果。     

局部投影可分离的高光谱图像异常检测

针对高光谱异常检测中临近异常像素相互干扰和背景地物复杂的问题,提出基于局部投影可分离的高光谱图像异常检测算法.在归一化的数据中,将待测像素光谱作为参考光谱,构造目标子空间,然后把邻域背景像素投影到该子空间,用投影后向量模值构造异常度计算式.最后将检测到的异常与全局主要背景地物进行比对,肖除部分虚警.利用HyMap高光谱数据进行仿真实验结果表明,本文算法具有克服背景复杂性和干扰点的影响,尤其对异类干扰点的抑制效果更佳.     

自适应不规则纹理的胶囊内镜图像无损压缩

为缓解无线胶囊内镜图像在电子设备以及服务器中的存储压力,提出一种自适应不规则纹理的无损压缩算法。在图像块内,利用扩展角度预测模式寻找与待预测像素最邻近的5个参考像素,并给其中3个参考像素分配不同权重,同时根据邻近像素值梯度变化规律,扩大待预测像素在不规则纹理方向上的预测值选择范围,基于图像块的最小信息熵选择最优的预测值,将真实值与预测值作差获得预测残差,以适应不规则纹理图像。利用跨分量预测模式选择最优的预测系数,构建符合图像块内预测残差分布规律的线性关系,从而消除当前编码像素中3个分量的冗余数据。结合Deflate算法对经多角度预测模式与跨分量预测模式预测后的剩余残差进行熵编码。实验结果表明,该算法在Kvasir-Capsule数据集上的无损压缩比平均为5.81,相比WebP、SAP、MDIP等算法,具有较优的压缩性能,能够有效提高图像的冗余消除率,其中相较WebP算法的冗余消除率提高约1.9%。     

A linear-time algorithm for Euclidean feature transform sets

The Euclidean distance transform of a binary image is the function that assigns to every pixel the Euclidean distance to the background. The Euclidean feature transform is the function that assigns to every pixel the set of background pixels with this distance. We present an algorithm to compute the exact Euclidean feature transform sets in linear time. The algorithm is applicable in arbitrary dimensions.     

Energy-based blob analysis for improving precision of skin segmentation

This paper addresses a problem of precise skin segmentation necessary for sign language recognition purposes. The main contribution of the presented research is an adaptive skin model enhanced with a blob analysis algorithm which significantly reduces false positives and improves skin segmentation precision. Adaptive skin detector utilizes a statistical skin color model updated dynamically based on a face region defined by eye positions. Face geometry is used for face and eye detection in luminance channel prior to the model adaptation. Color-based skin detectors classify every pixel separately which results in high false positives for background pixels which color is similar to human skin. The proposed blob analysis technique verifies detected skin regions by taking into account pixel topology. The experiments for ECU database showed that with the proposed approach false positive rate was reduced from 15.6% to 6% compared with a statistical model in RGB, which can be regarded as a significant improvement.     

An algorithm for recognizing linear objects in aerial photos automatically

A new algorithm for detecting linear infrastructural objects in aerial photos is presented. It is assumed that these objects pass through the whole image: beginning at one side and finishing at the opposite one. It is also assumed that the altitude of shooting and the image scale are invariable. The presented algorithm synthesizes the operation of an edge detector, a ridge detector, and the Hough accumulator into an object-of-interest mask excluding lots of spurious responses, and it completes missing information missed by detectors. First of all the image is preprocessed and anisotropically and repeatedly shrunk along the direction of the linear object and synthesis is performed by finding the shortest paths in a graph. The graph is presented in the form of a mesh, where each mesh node corresponds to a pixel of the shrunk image. At each node on the edges and in ridge lines, its energy is calculated, which is the reliability of this pixel. Then, the path that maximizes the sum of energies at the nodes is determined by considering its curvature. The obtained paths form a mask of linear objects. The algorithm is verified by using aerial photos for different seasons, and it demonstrates proper results (accuracy is ~80%).     

A spectral-spatial method based on low-rank and sparse matrix decomposition for hyperspectral anomaly detection

Recently, some methods based on low-rank and sparse matrix decomposition (LRASMD) have been developed to improve the performance of hyperspectral anomaly detection (AD). However, these methods mainly take advantage of the spectral information in hyperspectral imagery (HSI), and ignore the spatial information. This article proposes an LRASMD-based spectral-spatial (LS-SS) method for hyperspectral AD. First, the Go Decomposition (GoDec) algorithm is employed to solve the low-rank background component and the sparse anomaly component. Next, the sparse component is explored to calculate the spectral sparsity divergence index (SDI). Based on spectral SDI, the detection result in the spectral domain and the reliable background points, which are employed as training data to construct the background manifold by linear local tangent space alignment (LLTSA), can also be obtained. Then, based on the background manifold and the transformation matrix, the low-dimensional manifold of the whole data is computed by linear mapping. After that, the kernel collaborative representation detector (KCRD) is used in the low-dimensional manifold of the whole data for the spatial SDI. Finally, SS SDI is computed for the final detection result. The theoretical analysis and experimental results demonstrate that the proposed LS-SS can achieve better performance when compared with the comparison algorithms.     

压缩感知在Micro-CT图像超分辨重建中的应用

Micro-CT成像中重建图像的分辨率往往受到X射线的辐射剂量和探测器单元的孔径及大小的限制。在不改变原有成像参数的前提下,通过将重建图像网格的上采样以及重建图像的稀疏性假设先验,提出一种基于全变差模型的Micro-CT图像超分辨率重建模型。基于扩展梯度投影方法,将模型解耦分解为沿保真项的梯度方向下降、TV去噪、两步迭代结果线性组合这3步交替迭代求解。对模拟图像和实际数据进行了仿真测试,并同传统的滤波反投影方法进行了比较。实验结果表明,该算法能够有效提高重建图像的分辨率。     

Robust regression based face recognition with fast outlier removal

In this paper, we propose a new robust face recognition method through pixel selection. The method is based on the subspace assumption that a face can be represented by a linear combination in terms of the samples from the same subject. In order to obtain a reliable representation, only a subset of pixels with respect to smallest residuals are taken into the estimation. Outlying pixels which deviate from the linear model of the majority are removed using a robust estimation technique — least trimmed squares regression (LTS). By this method, the representation residual with each class is computed from only the clean data, which gives a more discriminant classification rule. The proposed algorithm provides a novel way to tackle the crucial occlusion problem in face recognition. Evaluation of the proposed algorithm is conducted on several public databases for the cases of both artificial and nature occlusions. The promising results show its efficacy.     

RX architectures for real-time anomaly detection in hyperspectral images

In the field of hyperspectral image processing, anomaly detection (AD) is a deeply investigated task whose goal is to find objects in the image that are anomalous with respect to the background. In many operational scenarios, detection, classification and identification of anomalous spectral pixels have to be performed in real time to quickly furnish information for decision-making. In this framework, many studies concern the design of computationally efficient AD algorithms for hyperspectral images in order to assure real-time or nearly real-time processing. In this work, a sub-class of anomaly detection algorithms is considered, i.e., those algorithms aimed at detecting small rare objects that are anomalous with respect to their local background. Among such techniques, one of the most established is the Reed–Xiaoli (RX) algorithm, which is based on a local Gaussian assumption for background clutter and locally estimates its parameters by means of the pixels inside a window around the pixel under test (PUT). In the literature, the RX decision rule has been employed to develop computationally efficient algorithms tested in real-time systems. Initially, a recursive block-based parameter estimation procedure was adopted that makes the RX processing and the detection performance differ from those of the original RX. More recently, an update strategy has been proposed which relies on a line-by-line processing without altering the RX detection statistic. In this work, the above-mentioned RX real-time oriented techniques have been improved using a linear algebra-based strategy to efficiently update the inverse covariance matrix thus avoiding its computation and inversion for each pixel of the hyperspectral image. The proposed strategy has been deeply discussed pointing out the benefits introduced on the two analyzed architectures in terms of overall number of elementary operations required. The results show the benefits of the new strategy with respect to the original architectures.     

一种基于主成分分析的 Codebook 背景建模算法

混合高斯(Mixture of Gaussian, MOG)背景建模算法和Codebook背景建模算法被广泛应用于监控视频的运动目标检测问题,但 混合高斯的球体模型通常假设RGB三个分量是独立的, Codebook的圆柱体模型假设背景像素值在圆柱体内均匀分布且背景亮度值变化方向指向坐标原点,这 些假设使得模型对背景的描述能力下降. 本文提出了一种椭球体背景模型,该模型克服了混合高斯球体模型和Codebook圆柱体模型假设的局限 性,同时利用主成分分析(Principal components analysis, PCA)方法来刻画椭球体背景模型, 提出了一种基于主成分分析的Codebook背景建模算法.实验表明,本文算法不仅能够更准确地描述背 景像素值在RGB空间中的分布特征,而且具有良好的鲁棒性.