Robust high-order matched filter for hyperspectral target detection with quasi-Newton method

Liu Liu; Zhenwei Shi; Shuo Yang; Haohan Zhang

doi:10.1109/CVRS.2012.6421234

Robust high-order matched filter for hyperspectral target detection with quasi-Newton method

Liu Liu^*
, Zhenwei Shi
, Shuo Yang
, Haohan Zhang

^*Corresponding author for this work

Beihang University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Robust high-order matched filter (RHMF), utilizing high-order statistics and considering the inherent variability in target spectral signatures, has obtained better results than other classical detection methods through experiments. However, this algorithm fails to get a fast convergence result by using simple steepest decent. In this paper, we accelerate this algorithm- RHMF successfully by introducing quasi-Newton method and DFP corrector formula, which is a more effective optimization algorithm based on second derivation, into this algorithm. We experiment constrained energy minimization (CEM), adaptive coherence estimator (ACE), RHMF with the steepest descent, and RHMF with quasi-Newton method on real data. The experiment by using RHMF with quasi-Newton has better and faster result, indicating that it is more effective for hyperspectral target detection. We also give the proof of the convergence of this method.

Original language	English
Title of host publication	Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012
Pages	63-66
Number of pages	4
DOIs	https://doi.org/10.1109/CVRS.2012.6421234
State	Published - 2012
Event	2012 International Conference on Computer Vision in Remote Sensing, CVRS 2012 - Xiamen, China Duration: 16 Dec 2012 → 18 Dec 2012

Publication series

Name	Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012

Conference

Conference	2012 International Conference on Computer Vision in Remote Sensing, CVRS 2012
Country/Territory	China
City	Xiamen
Period	16/12/12 → 18/12/12

Keywords

RHMF
hyperspectral target detection
quasi-Newton

Access to Document

10.1109/CVRS.2012.6421234

Cite this

Liu, L., Shi, Z., Yang, S., & Zhang, H. (2012). Robust high-order matched filter for hyperspectral target detection with quasi-Newton method. In Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012 (pp. 63-66). Article 6421234 (Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012). https://doi.org/10.1109/CVRS.2012.6421234

@inproceedings{cf3714f4e34f4f92a03daa386e3d563a,

title = "Robust high-order matched filter for hyperspectral target detection with quasi-Newton method",

abstract = "Robust high-order matched filter (RHMF), utilizing high-order statistics and considering the inherent variability in target spectral signatures, has obtained better results than other classical detection methods through experiments. However, this algorithm fails to get a fast convergence result by using simple steepest decent. In this paper, we accelerate this algorithm- RHMF successfully by introducing quasi-Newton method and DFP corrector formula, which is a more effective optimization algorithm based on second derivation, into this algorithm. We experiment constrained energy minimization (CEM), adaptive coherence estimator (ACE), RHMF with the steepest descent, and RHMF with quasi-Newton method on real data. The experiment by using RHMF with quasi-Newton has better and faster result, indicating that it is more effective for hyperspectral target detection. We also give the proof of the convergence of this method.",

keywords = "RHMF, hyperspectral target detection, quasi-Newton",

author = "Liu Liu and Zhenwei Shi and Shuo Yang and Haohan Zhang",

year = "2012",

doi = "10.1109/CVRS.2012.6421234",

language = "英语",

isbn = "9781467312738",

series = "Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012",

pages = "63--66",

booktitle = "Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012",

note = "2012 International Conference on Computer Vision in Remote Sensing, CVRS 2012 ; Conference date: 16-12-2012 Through 18-12-2012",

}

Liu, L , Shi, Z, Yang, S & Zhang, H 2012, Robust high-order matched filter for hyperspectral target detection with quasi-Newton method. in Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012., 6421234, Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012, pp. 63-66, 2012 International Conference on Computer Vision in Remote Sensing, CVRS 2012, Xiamen, China, 16/12/12. https://doi.org/10.1109/CVRS.2012.6421234

Robust high-order matched filter for hyperspectral target detection with quasi-Newton method. / Liu, Liu ; Shi, Zhenwei; Yang, Shuo et al.
Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012. 2012. p. 63-66 6421234 (Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Robust high-order matched filter for hyperspectral target detection with quasi-Newton method

AU - Liu, Liu

AU - Shi, Zhenwei

AU - Yang, Shuo

AU - Zhang, Haohan

PY - 2012

Y1 - 2012

N2 - Robust high-order matched filter (RHMF), utilizing high-order statistics and considering the inherent variability in target spectral signatures, has obtained better results than other classical detection methods through experiments. However, this algorithm fails to get a fast convergence result by using simple steepest decent. In this paper, we accelerate this algorithm- RHMF successfully by introducing quasi-Newton method and DFP corrector formula, which is a more effective optimization algorithm based on second derivation, into this algorithm. We experiment constrained energy minimization (CEM), adaptive coherence estimator (ACE), RHMF with the steepest descent, and RHMF with quasi-Newton method on real data. The experiment by using RHMF with quasi-Newton has better and faster result, indicating that it is more effective for hyperspectral target detection. We also give the proof of the convergence of this method.

AB - Robust high-order matched filter (RHMF), utilizing high-order statistics and considering the inherent variability in target spectral signatures, has obtained better results than other classical detection methods through experiments. However, this algorithm fails to get a fast convergence result by using simple steepest decent. In this paper, we accelerate this algorithm- RHMF successfully by introducing quasi-Newton method and DFP corrector formula, which is a more effective optimization algorithm based on second derivation, into this algorithm. We experiment constrained energy minimization (CEM), adaptive coherence estimator (ACE), RHMF with the steepest descent, and RHMF with quasi-Newton method on real data. The experiment by using RHMF with quasi-Newton has better and faster result, indicating that it is more effective for hyperspectral target detection. We also give the proof of the convergence of this method.

KW - RHMF

KW - hyperspectral target detection

KW - quasi-Newton

UR - https://www.scopus.com/pages/publications/84874421622

U2 - 10.1109/CVRS.2012.6421234

DO - 10.1109/CVRS.2012.6421234

M3 - 会议稿件

AN - SCOPUS:84874421622

SN - 9781467312738

T3 - Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012

SP - 63

EP - 66

BT - Proceedings of International Conference on Computer Vision in Remote Sensing, CVRS 2012

T2 - 2012 International Conference on Computer Vision in Remote Sensing, CVRS 2012

Y2 - 16 December 2012 through 18 December 2012

ER -

Robust high-order matched filter for hyperspectral target detection with quasi-Newton method

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this