Robust principal component analysis for micro-doppler based automatic target recognition

C. Clemente; A.W. Miller; J.J. Soraghan

Robust principal component analysis for micro-doppler based automatic target recognition

C. Clemente, A.W. Miller, J.J. Soraghan

Electronic And Electrical Engineering

Research output: Contribution to conference › Paper › peer-review

49 Downloads (Pure)

Abstract

Dealing with real data it is likely that it will exhibit the presence of unexpected observations within the data which can affect the correct reduction of the representative features of a target signature. For the speciffc case of micro-Doppler based classiffcation this problem can appear in the feature selection stage. To address this problem the Robust PCA based on the Minimum Covariance Determinant (MCD) estimator is introduced. The proposed technique showed to improve the overall classiffcation accuracy.

Original language	English
Number of pages	6
Publication status	Published - Oct 2013
Event	3rd IMA conference on Mathematics in Defence - Malvern, United Kingdom Duration: 24 Oct 2013 → …

Conference

Conference	3rd IMA conference on Mathematics in Defence
Country/Territory	United Kingdom
City	Malvern
Period	24/10/13 → …

Keywords

micro-doppler
automatic target recognition
target classification
radar detection

Access to Document

Clemente-etal-Robust-principle-component-analysis-for-micro-dopplerSubmitted manuscript, 177 KB

Cite this

@conference{8bf7e805acda4dafbca9c4268e2ff1eb,

title = "Robust principal component analysis for micro-doppler based automatic target recognition",

abstract = "Dealing with real data it is likely that it will exhibit the presence of unexpected observations within the data which can affect the correct reduction of the representative features of a target signature. For the speciffc case of micro-Doppler based classiffcation this problem can appear in the feature selection stage. To address this problem the Robust PCA based on the Minimum Covariance Determinant (MCD) estimator is introduced. The proposed technique showed to improve the overall classiffcation accuracy.",

keywords = "micro-doppler, automatic target recognition, target classification, radar detection",

author = "C. Clemente and A.W. Miller and J.J. Soraghan",

year = "2013",

month = oct,

language = "English",

note = "3rd IMA conference on Mathematics in Defence ; Conference date: 24-10-2013",

}

TY - CONF

T1 - Robust principal component analysis for micro-doppler based automatic target recognition

AU - Clemente, C.

AU - Miller, A.W.

AU - Soraghan, J.J.

PY - 2013/10

Y1 - 2013/10

N2 - Dealing with real data it is likely that it will exhibit the presence of unexpected observations within the data which can affect the correct reduction of the representative features of a target signature. For the speciffc case of micro-Doppler based classiffcation this problem can appear in the feature selection stage. To address this problem the Robust PCA based on the Minimum Covariance Determinant (MCD) estimator is introduced. The proposed technique showed to improve the overall classiffcation accuracy.

AB - Dealing with real data it is likely that it will exhibit the presence of unexpected observations within the data which can affect the correct reduction of the representative features of a target signature. For the speciffc case of micro-Doppler based classiffcation this problem can appear in the feature selection stage. To address this problem the Robust PCA based on the Minimum Covariance Determinant (MCD) estimator is introduced. The proposed technique showed to improve the overall classiffcation accuracy.

KW - micro-doppler

KW - automatic target recognition

KW - target classification

KW - radar detection

UR - http://www.ima.org.uk/conferences/conferences_calendar/3rd_mathematics_in_defence.cfm.html

M3 - Paper

T2 - 3rd IMA conference on Mathematics in Defence

Y2 - 24 October 2013

ER -

Robust principal component analysis for micro-doppler based automatic target recognition

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this