Kernel RX-algorithm: a nonlinear anomaly detector for hyperspectral imagery

Kwon, Heesung; Nasrabadi, Nasser M.

doi:10.1109/tgrs.2004.841487

Cited by 649 publications

(112 citation statements)

References 23 publications

Supporting

Mentioning

112

Contrasting

Order By: Relevance

“…Heesung and Nasrabadi in [21] proposed (KRX) a nonlinear version of the RXD based on its kernelization in the feature space in terms of kernels that implicitly compute dot products in the feature space. To estimate the kernel matrix, which is a Gram matrix; proprieties of the kernel principal component analysis are used.…”

Section: B Kernel Based Anomaly Detectorsmentioning

confidence: 99%

“…This supposition accentuates the False Alarm Rate (FAR) especially for high resolution images where the supposition of homogeneity seems to be inappropriate since the big diversity of existing materials. To decrease this fact, non linear models of the background 978-1-4799-7069-8/14/$31.00 2014 IEEE have been proposed with the kernel based anomaly detectors [21][22][23][24][25][26][27][28]. Other researches try to solve the anomaly detection problem with different techniques as the projection [31][32][33][34][35][36] and the segmentation [37][38][39][40][41][42][43][44][45][46].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Anomaly detection in hyperspectral imagery: an overview

Salem

Ettabaa

Hamdi

2014

International Image Processing, Applications and Systems Conference

View full text Add to dashboard Cite

Interest on anomaly detection for hyperspectral images is increasingly growing the last decades due to the diversity of applications that aims for detecting small distinctive objects dispersed in a large geographic zone, without any prior knowledge about the scene. In addition to the absence of prior knowledge, many problems are particularly challenging for the anomaly detection such as the differentiation between real targets, false alarms and noise, the detection of anomaly of different shapes and sizes, and the high computational cost of the proposed approaches. This overview presents the literature and approaches proposed to address these issues. These approaches are grouped into four categories based on the underlying techniques used to achieve the detection: 1) statistical methods, 2) kernel based methods, 3) projection based methods and 4) segmentation based methods.

show abstract

Section: B Kernel Based Anomaly Detectorsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Anomaly detection in hyperspectral imagery: an overview

Salem

Ettabaa

Hamdi

2014

International Image Processing, Applications and Systems Conference

View full text Add to dashboard Cite

show abstract

“…To reduce the background pixels in the input data, we used the assumption in [3,19] that the input data consists of two Gaussian distributions, background distribution and target distribution. In other words, the target pixels are drawn from multivariate Gaussian distribution.…”

Section: A Data Reductionmentioning

confidence: 99%

“…Reducing the background pixels facilitate the detection of targets based on the entire input information of the targets and simultaneously reduce the computational burden on SEM algorithm. The reduction step is accomplished in the feature space where the input data is mapped using the Gaussian RBF kernel [19,20]. This fusion scheme of the SEM algorithm with kernel approach reduced the computational burden on the SEM algorithm, used the entire information of target for detection, and reduced the possibility of false alarms that may arise from noisy input data.…”

Section: Introductionmentioning

confidence: 99%

Kernel and stochastic expectation maximization fusion for target detection in hyperspectral imagery

Elbakary

Alam

2011

Optical Pattern Recognition XXII

View full text Add to dashboard Cite

In this paper, we present a new algorithm for target detection using hyperspectral imagery. The proposed algorithm is inspired by the outstanding performance of nonlinear RX-algorithm and the robustness of the stochastic expectation maximization (SEM) algorithm. The traditional technique of using SEM algorithm for target detection in hyperspectral imagery is associated with dimensionality reduction of the input data using binning or principal components analysis (PCA) algorithm. Although, the data reduction of the input data is enforced to reduce the computational burden on SEM algorithm, but it affects the results of target detection, especially the challenging one, due to not using the entire information of the potential targets. To facilitate detection of the target by using the entire targets information and simultaneously reducing the computational burden on SEM algorithm, we propose a new scheme for data reduction based on using Kernels. Kernel-based input data reduction is a nonlinear filtering technique in which the input data are mapped to the feature space where most of the background data is filtered using an easily selected threshold. Then, Gaussian mixture model is generated for the reduced input-data and SEM algorithm is employed to estimate the model parameters and to classify that input data. Finally, we allocated the target's class and isolated the target pixels. The proposed scheme for fusion the kernel with SEM algorithm has been tested using real life hyperspectral imagery and the results show superior performance compared to alternate algorithms.

show abstract

“…In most literatures of hyperspectral anomaly detection, for the sake of extruding the anomalies, background estimation is the most important issue (4)- (6) . A variety of anomaly detection algorithms, such as Reed-Xiaoli (RX) detector (7) , kernel-RX (KRX) (8) , and the cluster-based anomaly detector (CBAD) (9) , aim at suppressing the background land-covers via a statistical background estimation. However, these statistical based anomaly detection methods are limited in that they incorporate the anomaly targets in the background estimation when undertaking a global or clustered background statistics estimation.…”

Section: Introductionmentioning

confidence: 99%

Sparsity Score Estimation for Hyperspectral Anomaly Detection

Zhao

Han

et al. 2016

Proceedings of the 4th IIAE International Conference on Intelligent Systems and Image Processing 2016

View full text Add to dashboard Cite

Hyperspectral image usually possesses complicated conditions of land-cover distribution, which brings challenge to achieve an effective background representation for hyperspectral anomaly detection. Sparse learning gives a way to overcome this obstacle. A novel sparsity score estimation framework for hyperspectral anomaly detection (SSEAD) is proposed in this paper. Firstly, an overcomplete dictionary and corresponding sparse code matrix are obtained from the hyperspectral data. Then, the frequency of each dictionary atom for reconstruction, which is also called the atom usage probability (AUP), are estimated from the sparse code matrix, from which the strength of each atom for reconstruction is obtained. Finally, the estimated frequencies are transformed to the sparsity score for each pixel. In the proposed detection framework, two operations which aim to enhance the learned dictionary to be more effective for anomaly detection are implemented: 1) dictionary based background feature transformation, and 2) dictionary iterative reweighting. Two real-world hyperspectral datasets are utilized to evaluate the performance of the proposed framework. The experimental results show that the proposed framework achieves superior performance relative to some of the other state-of-the-art anomaly detection methods.

show abstract

Kernel RX-algorithm: a nonlinear anomaly detector for hyperspectral imagery

Cited by 649 publications

References 23 publications

Anomaly detection in hyperspectral imagery: an overview

Anomaly detection in hyperspectral imagery: an overview

Kernel and stochastic expectation maximization fusion for target detection in hyperspectral imagery

Sparsity Score Estimation for Hyperspectral Anomaly Detection

Contact Info

Product

Resources

About