Vous êtes sur la page 1sur 11

Remote Sensing Letters

ISSN: 2150-704X (Print) 2150-7058 (Online) Journal homepage: http://www.tandfonline.com/loi/trsl20

A novel anomaly detection method incorporating


target information derived from hyperspectral
imagery

Qiandong Guo, Ruiliang Pu, Lianru Gao & Bing Zhang

To cite this article: Qiandong Guo, Ruiliang Pu, Lianru Gao & Bing Zhang (2016) A novel
anomaly detection method incorporating target information derived from hyperspectral
imagery, Remote Sensing Letters, 7:1, 11-20, DOI: 10.1080/2150704X.2015.1101177

To link to this article: http://dx.doi.org/10.1080/2150704X.2015.1101177

Published online: 19 Oct 2015.

Submit your article to this journal

Article views: 32

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at


http://www.tandfonline.com/action/journalInformation?journalCode=trsl20

Download by: [Wuhan University] Date: 10 November 2015, At: 06:46


REMOTE SENSING LETTERS, 2016
VOL. 7, NO. 1, 1120
http://dx.doi.org/10.1080/2150704X.2015.1101177

A novel anomaly detection method incorporating target


information derived from hyperspectral imagery
Qiandong Guoa, Ruiliang Pua, Lianru Gaob and Bing Zhangb
a
School of Geosciences, University of South Florida, Tampa, FL, USA; bKey Laboratory of Digital Earth
Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing, China

ABSTRACT ARTICLE HISTORY


Anomaly detection is an active research topic in hyperspectral Received 5 March 2015
remote sensing and has been applied in many areas including Accepted 22 September
Downloaded by [Wuhan University] at 06:46 10 November 2015

environmental monitoring, urban survey, mineral mapping, and 2015


national security. Usually, it makes a detection decision without
any prior target information or background information. Several
anomaly detection algorithms (e.g. ReedXiaoli detector, blocked
adaptive computationally ecient outlier nominator and random-
selection-based anomaly detector) have been developed, which
rely on estimating background information only from a hyperspec-
tral image without considering target information in making a
detection decision. These methods may be ecient in general
but sometimes with high false alarm rate (FAR). In order to reduce
FAR, this study proposes a novel method that incorporates both
background and target information, derived from the hyperspec-
tral imagery, into anomaly detection algorithms. The target infor-
mation is helpful to detect anomalies as outliers. With a scene of
real airborne visible infrared imaging spectrometer data, the
experimental results demonstrate that the proposed method has
produced better detection results and higher time eciency com-
pared with those using the traditional algorithms that only con-
sider background information.

1. Introduction
Anomaly detection is one of the important topics in hyperspectral image processing
and hyperspectral remote sensing applications. Anomaly can be dened as a target
which appears with a low probability in a given scene (Eismann 2012). Although no
targets or background information is available in the process of detection, anomalies
still have two characteristics that make them as outliers: (1) their spectral signatures
are dierent from their neighbour pixels and (2) anomalies occur with a low prob-
ability (Stein et al. 2002). Based on these two characteristics, statistical models have
been developed and used to calculate the probability of a pixel under test (PUT) being
a target. One main assumption is that its background follows a multivariate normal
distribution. According to this assumption, ReedXiaoli detector (RXD) (Reed and Yu
1990) was developed and widely used for anomaly detection. It applies the probability

CONTACT Bing Zhang zb@radi.ac.cn


2015 Taylor & Francis
12 Q. GUO ET AL.

density function of a multivariate normal distribution to calculate the probability of a


PUT of being a part of the background. However, at least two factors inuence the
performance of this detector. (1) The assumption held by RXD in which the back-
ground is a multivariate normal distribution is simple for many real scenarios.
Sometimes, a scene contains a variety of objects that are too complex to be consid-
ered as a multivariate normal distribution (Huck and Guillaume 2010). (2) Anomaly
detection is a condition without target spectral information. It is a blind detecting
process which may cause some uninterested anomalies of being detected as targets.
These factors complicate anomaly detection tasks and increase the error of detecting
background pixels as targets, thus increasing the false alarm rate (FAR) of RXD.
A variety of strategies have been implemented to improve the eciency of RXD.
Several methods focus on how to make background closer to multivariate normal
distribution. They rene the background by removing anomalies or reducing the weight
of anomalies in the background samples. These algorithms include the local-RXD (LRXD)
Downloaded by [Wuhan University] at 06:46 10 November 2015

(Gorelnik, Yehudai, and Rotman 2010), the blocked adaptive computationally ecient
outlier nominator (BACON) (Billor, Hadi, and Velleman 2000), the random-selection-
based anomaly detector (RSAD) (Du and Zhang 2011), the weighted-RXD (Guo et al.
2014) and a discriminative metric learning-based anomaly detection method (Du and
Zhang 2014). In recent years, several other approaches such as a feature extraction
algorithm named sparse transfer manifold embedding and a supervised matric learning
algorithm are implemented for subpixel target detection (Zhang, Zhang, Tao, and Huang
2014; Zhang, Zhang, Tao, Huang, et al. 2014). BACON and RSAD are two novel algo-
rithms among these detectors. The two algorithms can prevent the contamination by
removing anomalous signatures when estimating background information. They are
very powerful to spot anomalies as outliers. However, it is noticeable that the two
methods can be used to estimate information from the background only. As mentioned
earlier, lacking target information could be a drawback that might cause a high FAR.
Usually, a scene of hyperspectral imagery can also contain several targets which can be
spotted by the RXD algorithm. For this case, we propose a new method that can be used
not only to estimate background information, but also to calculate target statistics from
the image. The additional anomaly information can be used to improve the ability of
detectors to spot anomalies.

2. Multivariate normal distribution model for anomaly detection


2.1. RXD
RXD assumes that the background in a hyperspectral image follows a multivariate
normal distribution, which can be described as follows. Let H1 be the target signal
and H0 be the background signal. The detection problem can be dened as
H0 : x b (1)

H1 : x s b (2)

where x is the sample pixel vector; s is the target signal and b is the background signal
vector which is assumed as a multivariate normal distribution with mean vector and
REMOTE SENSING LETTERS 13

covariance matrix S, i.e. b is modelled as N; S, a multivariate normal distribution


whose mean vector is and covariance matrix is S. Therefore, xjH0 are modelled as
N; S and x j H1 are modelled as N s; S. Based on statistic knowledge, we can
obtain the probability of x j H0 :

1 T 1
e2x S x
1
Px j H0 K=2 1=2
(3)
2 jSj

where K is the number of bands of a hyperspectral image. The symbol T means


transposing the matrix. Since an anomaly x s is expected to be signicantly dierent
from the background in a spectral space, Px s j H0 should be very small for an
anomalous pixel. Therefore, for a given background, as 1
is xed,
T 1 2K=2 jSj1=2
x s  S x s  should be larger for an anomalous pixel than for a background
Downloaded by [Wuhan University] at 06:46 10 November 2015

pixel. Based on this observation, RXD uses the following expression to detect anomalies:

RXDx x  T S1 x  (4)

2.2. BACON
BACON is an algorithm to spot outliers in multivariate and regression data. It uses two
main strategies to achieve the goal: (1) thresholding RXD values of pixels to rene the
background and (2) selecting a small group of background samples to keep the time
eciency. The BACON algorithm includes the following steps:

Step 1. Compute the RXD score for each PUT. Select m cK smallest values of RXD
scores as the initial background subset, where K is the number of bands; cis a small
integer chosen by the data analyst, and c should be greater than 1 such that the
condition of m > K can be met.
Step 2. Obtain the square root of RXD scores based on the current background subset.
Step 3. Those pixels whose square root values of RXD scores are smaller than cn;K;r K;
will be selected as new background samples, where K; is the square root of the
1  percentile of the x2 distribution with K degrees of freedom, and cn;K;r is
computed as follows:

cn;K;r cn;K ch;r (5)

ch;r maxf0; h  r=h rg (6)

h n K 1=2 (7)

cn;K 1 K 1=n  K 1=n  h  K (8)

where n is the total number of pixels and r is the number of pixels in current back-
ground subset.

Step 4. Iterate steps 2 and 3 until the size of the background subset no longer change.
Step 5. Map anomalies to the image space.
14 Q. GUO ET AL.

2.3. RSAD
Dierent from BACON, the RSAD algorithm randomly selects representative background
samples from the image to estimate background statistical information, identies
anomalies by statistical dierences, and nally, fuses all the detection results together.
The steps of RSAD are described as follows:

Step 1. Randomly select background pixels as the initial background subset of


observed pixels from a hyperspectral image.
Step 2. Compute the square root of RXD value of each pixel vector based on current
background subset.
Step 3. Those pixels whose square root values of RXD scores are smaller than cn;K;r K;
will be selected as new background samples pixels. The procedures to compute
cn;K;r are the same as those for BACON.
Downloaded by [Wuhan University] at 06:46 10 November 2015

Step 4. Iterate steps 2 and 3 until the background subset no longer change.
Step 5. Label the pixels excluded by the nal background subset as anomalies.

3. Anomaly detectors incorporating target information


As discussed above, BACON and RSAD only select a subgroup as a background sample
set for estimating the background of anomalies. They do not try to use potential target
information to assist anomaly detection. In this study, we propose a method to extract
both background and target information for anomaly detection. As demonstrated in
Section 2, RXD assumes that the target signal is a constant vector s. Based on
the Equations (1) and (2), the data under H0 are modelled as N; S and the data
under H1 are modelled as N s; S (Gu, Liu, and Zhang 2008). Thus, the covariance
matrices of the background and target are identical. Unfortunately, in most images,
there are several dierent types of targets, which may violate the assumption of a
constant vector s so that the representation of the targets signals is inappropriate. It
may be more reasonable that the target signals are expressed as variable vectors and
have dierent covariance matrices, which are dierent from the covariance matrix of the
background (Gao et al. 2014). As BACON selects the pixels with the smallest values of
RXD scores into background subset, our method chooses the pixels with the largest
values of BACON or RSAD scores as target samples. Based on this principle, we can
obtain the target set V1 and the background set V0. The both background and target
sets can be assumed as multivariate normal distribution (Gao et al. 2014). Therefore, the
posterior probabilities PHm j x are obtained as follows:
1 T
S1
e2xm m xm
1
Px j Hm K=2 1=2
m 0; 1 (9)
2 j Sm j

Nm
PHm (10)
N0 N1
Therefore,
PHm j x Px j Hm  PHm =Px (11)
REMOTE SENSING LETTERS 15

where Px is the probability that vector x occurs in the image, Sm is the covariance
matrix of Vm; m is the mean vector of Vm and Nm is the number of samples in Vm. In this
case, we can use the log-likelihood ratio to simplify the equation:

1 K
lnPHm j x  x  m T S1
m x  m  ln2
2 2 (12)
1
 lnj Sm j lnPHm   lnPx
2

Then subtracting lnPH0 j x from lnPH1 j x, we obtain the quantity d10 :

1 1
d10 x  0 T S1 T 1
0 x  0  x  1 S1 x  1
2 2 (13)
1 1
lnj S0 j lnPH1   lnj S1 j  lnPH0 
2 2
Downloaded by [Wuhan University] at 06:46 10 November 2015

As 12 lnj S0 j lnPH1   12 lnjS1 j  lnPH0  is constant for a considered image,


d10 can be simplied as

d10 x  0 T S1 T 1
0 x  0  x  1 S1 x  1 RXD0 x  RXD1 x (14)

Resulting from the previous equations, RXD0 x represents the distance from the
PUTx to the background set, and RXD1 x is the distance from x to the target set.
Although the algorithm of Equation (14) is not a constant FAR detector (Gao et al. 2014),
it still has three advantages: (1) this algorithm distinguishes the covariance matrix of the
background from the covariance matrix of targets. The additional target information can
benet for the target detection; (2) this algorithm highlights an anomaly twice by the
distance from the background set and by the distance from the target set; and (3)
background pixels are signicantly dierent from targets in spectral domain. Thus, the
RXD1 x of the background pixels will be very large to withhold them as outliers. As the
number of target samples chosen by the proposed method is determined by the
number of bands, it is necessary to control the size of target sample, otherwise more
background information may be involved in the target set. Since principal component
analysis (PCA) is a commonly used approach to reduce data dimensionality and increase
the detection eciency (Ma, Crawford, and Tian 2010; Taitano, Geier, and Bauer 2010; Du
and Zhang 2011), our method uses it to control the number of target samples while
preserving major information of the image. The procedures of our method can be
described as follows:

Step 1. Conduct PCA transformation for an image and keep the rst t principal
component (PC) images.
Step 2. Implement BACON or RSAD with the rst t component images to create
background subset.
Step 3. Select l ct pixels with the largest values of BACON scores or RSAD scores
into target subset, where c is the small integer for determining background
samples in BACON or RSAD.
Step 4. Using the background subset and target subset created by steps 2 and 3,
based on Equation (14) to produce a grey image as nal output.
16 Q. GUO ET AL.

4. Experiments with real hyperspectral image data


4.1. Hyperspectral data set
In this section, one real data set has been used for experimental evaluation of the
proposed methods. The data set was collected by the airborne visible infra-red
imaging spectrometer (AVIRIS), operated by NASAs Jet Propulsion Laboratory, over
the World Trade Center (WTC) area in New York, on 16 September 2001 (just ve days
after the terrorist attacks that collapsed the two main towers in the WTC complex)
(Molero et al. 2013). A portion of 200  200 pixels (with 224 spectral bands between
0.4 and 2.5 m) was selected for the test in this study. The data set covered some hot
spots corresponding to latent res at the WTC, which can be considered as anomalies.
Figure 1(a) shows a false colour composite image of AVIRIS data selected for the
experiment, while Figure 1(b) displays a ground truth data, which comprises spatial
Downloaded by [Wuhan University] at 06:46 10 November 2015

locations of the hot spots provided by the United States Geological Survey. The
ground truth image was used to evaluate the performances of dierent algorithms.

4.2. Analysis of ROC and AUC


In this study, we rst used the BACON and RSAD algorithms to obtain target sets and
calculate their corresponding statistic information. The new methods were denoted as
BACON-Target and RSAD-Target, based on which result we used. Since the experimental
results have demonstrated that BACON performed quite well when c 4 or 5 (Billor,
Hadi, and Velleman 2000), we set c 4 for this experiment. We then compared the new
methods with the original BACON and RSAD algorithms to demonstrate whether there
were any improvements with our new methods. There were at least two criteria that
might be used to evaluate the performances of detection algorithms: (1) receiver
operating characteristic (ROC) curve (Gaucel, Guillaume, and Bourennane 2008; Qian,
Levesque, and Far 2011; Eismann 2012; Molero et al. 2012) and (2) the area under the
ROC curves (AUC) (Khazai et al. 2011). Given the dierent strategies to obtain samples of
background, global-RXD (GRXD) and LRXD were also used for this data set (Gorelnik,
Yehudai, and Rotman 2010). Specically, since the size of target sample in an image is

Figure 1. (a) AVIRIS image over the World Trade Center in New York City on 16 September 2001; (b)
ground truth map indicating spatial locations of the hot spot res, available from the United States
Geological Survey.
REMOTE SENSING LETTERS 17

Figure 2. Detection results created by dierent algorithms on the WTC data.


Downloaded by [Wuhan University] at 06:46 10 November 2015

usually more than one pixel, the LRXD in this experiment had been implemented using a
dual window technique (Kwon, Der, and Nasrabadi 2003). We used dierent inner and
outer window sizes to process the image. 15  15as an outer window size and 11  11as
an inner window size were selected for this experiment as they led to the best detection
performance on this data set. The grey images obtained by the mentioned detectors are
presented in Figure 2. We could identify the re spots by their high brightness values.
The AUCs of dierent detectors for various PCs are shown in Figure 3. It is noticeable
that BACON and RSAD would produce the best detection results when using 15 PCs.
Thus, BACON-Target (15 PCs) and RSAD-Target (15 PCs) would be implemented in the
following experiments. The spectral information was inadequate if fewer components
were used. On the other hand, more PCs could probably contain more noise, which
could negatively impact the capability of detectors. Thus, either using too few or too

Figure 3. AUCs created by dierent algorithms by using dierent numbers of PCs on WTC data.
18 Q. GUO ET AL.

many components would lead to the reduction of accuracy. BACON-Target and RSAD-
Target outperformed BACON and RSAD on the rst 5 PCs to rst 17 PCs, which demon-
strated that target information could improve the capability of these anomaly detectors.
GRXD created a better result than LRXD, probably because some re spots in the study
area were large in size, and thus the local windows always included some anomaly pixels
when estimating information of background, which withheld the anomalies to be
detected.
All ROCs of the discussed methods are presented in Figure 4, and their correspond-
ing AUCs are shown in Table 1. From the gure, RSAD-Target (15 PCs) had the best
detection performance among all these detectors. BACON and RSAD performed with
almost no dierence when full data or 15 PCs were used. However, the computation
time of BACON (full) was 11.82 times higher than that by BACON (15 PCs). Thus, it is
wise to use several PCs instead of full bands of the image when implementing
BACON or RSAD. GRXD was better than LRXD in spite of using full bands or 15 PCs.
Downloaded by [Wuhan University] at 06:46 10 November 2015

In terms of time eciency, GRXD was the fastest algorithm as it was the simplest.
However, it created more false alarms than BACON, RSAD and the proposed methods.
BACON-Target (15 PCs) and RSAD-Target (15 PCs) outperformed BACON and RSAD in
spite of using 15 PCs or full bands of the image, which demonstrated that the target
information could help detectors spot more targets with less false alarms indeed.

Figure 4. ROC curves corresponding to the detection results presented in Figure 2.

Table 1. AUCs and processing times (seconds) produced by dierent detectors; also see Figures 2
and 4.
Algorithms AUC Time (s) Algorithms AUC Time (s)
GRXD (full) 0.9689 9.11 LRXD (full) 0.9642 36.10
GRXD (15 PCs) 0.9583 0.29 LRXD (15 PCs) 0.9082 13.08
BACON (full) 0.9947 97.03 RSAD (full) 0.9946 49.55
BACON (15 PCs) 0.9942 8.21 RSAD (15 PCs) 0.9944 8.92
BACON-Target (15 PCs) 0.9978 8.34 RSAD-Target (15 PCs) 0.9987 9.04
REMOTE SENSING LETTERS 19

Several recent papers mentioned using kernel methods to detect anomalies (e.g.
Kernel RX algorithm and Kernel PCA). These algorithms could mine the high-order
correlation between spectral bands and selects the non-linear features from the
hyperspectral image (Kwon and Nasrabadi 2005; Gu, Liu, and Zhang 2008).
However, kernel methods are greatly time consuming and need a lot of memory
space. Compared with these methods, our proposed methods are much easier to
implement.

5. Conclusion
Anomaly detection is a very active topic of research in hyperspectral imagery processing.
Usually, it is with a condition without any prior background or target information.
Currently existing algorithms (e.g. RXD, BACON and RSAD) only estimate the information
Downloaded by [Wuhan University] at 06:46 10 November 2015

of background to make decisions for anomaly detection. This study proposed new
methods of incorporating target information from the image to assist detection. The
new methods were combined with PCA technique in order to control the number of
target samples as well as increase the time eciency. In this study, we used real
hyperspectral data (AVIRIS) to test the performances of our new methods. The experi-
mental results demonstrate that the new methods have outperformed other traditional
algorithms which only consider background information (e.g. BACON, RSAD and RXD).
Moreover, the proposed methods were of time eciency as they could obtain satised
detection results with only a few PC images.

Acknowledgement
The authors would like to express appreciation to the NASAs Jet Propulsion Laboratory for
providing the WTC data set.

Disclosure statement
No potential conict of interest was reported by the authors.

References
Billor, N., A. S. Hadi, and P. F. Velleman. 2000. BACON: Blocked Adaptive Computationally Ecient
Outlier Nominators. Computational Statistics & Data Analysis 34 (3): 279298. doi:10.1016/
S0167-9473(99)00101-2.
Du, B., and L. Zhang. 2011. Random-Selection-Based Anomaly Detector for Hyperspectral
Imagery. IEEE Transactions on Geoscience and Remote Sensing 49 (5): 15781589. doi:10.1109/
TGRS.2010.2081677.
Du, B., and L. Zhang. 2014. A Discriminative Metric Learning Based Anomaly Detection Method.
IEEE Transactions on Geoscience and Remote Sensing 52 (11): 68446857. doi:10.1109/
TGRS.2014.2303895.
Eismann, M. 2012. Hyperspectral Remote Sensing. Bellingham, WA: SPIE Press. doi:10.1117/3.899758.
Gao, L., Q. Guo, A. Plaza, J. Li, and B. Zhang. 2014. Probabilistic Anomaly Detector for Remotely
Sensed Hyperspectral Data. Journal of Applied Remote Sensing 8 (1): 083538083538.
doi:10.1117/1.JRS.8.083538.
20 Q. GUO ET AL.

Gaucel, J.-M., M. Guillaume, and S. Bourennane. 2008. Adaptive Linear Minimum Mean Square
Error Restoration: Inuence on Hyperspectral Detection Strategy. International Journal of
Remote Sensing 29 (10): 29432961. doi:10.1080/01431160701408428.
Gorelnik, N., H. Yehudai, and S. R. Rotman. 2010. Anomaly Detection in Non-Stationary
Backgrounds. 2nd Workshop on Hyperspectral Image and Signal Processing: Evolution in
Remote Sensing, Reykjavik, Iceland: WHISPERS 2010, IEEE Computer Society, June 1416.
doi:10.1109/WHISPERS.2010.5594914.
Gu, Y., Y. Liu, and Y. Zhang. 2008. A Selective KPCA Algorithm Based on High-Order Statistics for
Anomaly Detection in Hyperspectral Imagery. IEEE Geoscience and Remote Sensing Letters 5 (1):
4347. doi:10.1109/LGRS.2007.907304.
Guo, Q., B. Zhang, Q. Ran, L. Gao, J. Li, and A. Plaza. 2014. Weighted-RXD and Linear Filter-Based
RXD: Improving Background Statistics Estimation for Anomaly Detection in Hyperspectral
Imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 7
(6): 23512366. doi:10.1109/JSTARS.2014.2302446.
Huck, A., and M. Guillaume. 2010. Asymptotically CFAR-Unsupervised Target Detection and
Discrimination in Hyperspectral Images With Anomalous-Component Pursuit. IEEE
Downloaded by [Wuhan University] at 06:46 10 November 2015

Transactions on Geoscience and Remote Sensing 48 (11): 39803991. doi:10.1109/


TGRS.2010.2063434.
Khazai, S., S. Homayouni, A. Safari, and B. Mojaradi. 2011. Anomaly Detection in Hyperspectral
Images Based on an Adaptive Support Vector Method. IEEE Geoscience and Remote Sensing
Letters 8 (4): 646650. doi:10.1109/LGRS.2010.2098842.
Kwon, H., S. Z. Der, and N. M. Nasrabadi. 2003. Adaptive Anomaly Detection Using Subspace
Separation for Hyperspectral Imagery. Optical Engineering 42 (11): 33423351. doi:10.1117/
1.1614265.
Kwon, H., and N. M. Nasrabadi. 2005. Kernel Rx-Algorithm: A Nonlinear Anomaly Detector for
Hyperspectral Imagery. IEEE Transactions on Geoscience and Remote Sensing 43 (2): 388397.
doi:10.1109/TGRS.2004.841487.
Ma, L., M. M. Crawford, and J. Tian. 2010. Anomaly Detection for Hyperspectral Images Based on
Robust Locally Linear Embedding. Journal of Infrared, Millimeter, and Terahertz Waves 31 (6):
753762. doi:10.1007/s10762-010-9630-3.
Molero, J. M., E. M. Garzon, I. Garcia, and A. Plaza. 2012. Anomaly Detection Based on a Parallel
Kernel RX Algorithm for Multicore Platforms. Journal of Applied Remote Sensing 6 (1): 061503
106150310. doi:10.1117/1.JRS.6.061503.
Molero, J. M., E. M. Garzon, I. Garcia, and A. Plaza. 2013. Analysis and Optimizations of Global and
Local Versions of the RX Algorithm for Anomaly Detection in Hyperspectral Data. IEEE Journal
of Selected Topics in Applied Earth Observations and Remote Sensing 6 (2): 801814. doi:10.1109/
JSTARS.2013.2238609.
Qian, S. E., J. Levesque, and R. R. Far. 2011. Assessment of Noise Reduction of Hyperspectral
Imagery Using a Target Detection Application. International Journal of Remote Sensing 32 (12):
32673284. doi:10.1080/01431161003745640.
Reed, I. S., and X. Yu. 1990. Adaptive Multiple-Band CFAR Detection of an Optical Pattern with
Unknown Spectral Distribution. IEEE Transactions on Acoustics, Speech and Signal Processing 38
(10): 17601770. doi:10.1109/29.60107.
Stein, D. W. J., S. G. Beaven, L. E. Ho, E. M. Winter, A. P. Schaum, and A. D. Stocker. 2002. Anomaly
Detection from Hyperspectral Imagery. IEEE Signal Processing Magazine 19 (1): 5869.
doi:10.1109/79.974730.
Taitano, Y., B. Geier, and K. Bauer. 2010. A Locally Adaptable Iterative RX Detector. EURASIP
Journal on Advances in Signal Processing 2010: 341908. doi:10.1155/2010/341908.
Zhang, L., L. Zhang, D. Tao, and X. Huang. 2014. Sparse Transfer Manifold Embedding for
Hyperspectral Target Detection. IEEE Transactions on Geoscience and Remote Sensing 52 (2):
10301043. doi:10.1109/TGRS.2013.2246837.
Zhang, L., L. Zhang, D. Tao, X. Huang, and B. Du. 2014. Hyperspectral Remote Sensing Image
Subpixel Target Detection Based on Supervised Metric Learning. IEEE Transactions on
Geoscience and Remote Sensing 52 (8): 49554965. doi:10.1109/TGRS.2013.2286195.

Vous aimerez peut-être aussi