A spectral-spatial hyperspectral data classification approach using random forest with label constraints

Yuemei Ren; Yanning Zhang; Wei Wei; Lei Li

doi:10.1109/IWECA.2014.6845627

A spectral-spatial hyperspectral data classification approach using random forest with label constraints

Yuemei Ren
, Yanning Zhang
, Wei Wei
, Lei Li

School of Computer Science

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

25 Scopus citations

Abstract

A new classification approach using random forest with label constraints is proposed to deal with the underutilization effectively of spectral and spatial information for hyperspectral image classification. Firstly, the principal component analysis extraction method is adopted, and the extended morphological profiles of the image are extracted from the principle components images by using mathematical morphology method. Then random forest is constructed based on the extracted features. Finally, the label constraints based on space continuity is used to constraint the results by using the label information of its neighborhoods on image space. The classification result is decided by voting strategy. Experimental results of several real hyperspectral images demonstrate that the proposed approach outperforms the random forest method without constraint and the popular SVM classification method.

Original language	English
Title of host publication	Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014
Publisher	IEEE Computer Society
Pages	344-347
Number of pages	4
ISBN (Print)	9781479945658
DOIs	https://doi.org/10.1109/IWECA.2014.6845627
State	Published - 2014
Event	2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014 - Ottawa, ON, Canada Duration: 8 May 2014 → 9 May 2014

Publication series

Name	Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014

Conference

Conference	2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014
Country/Territory	Canada
City	Ottawa, ON
Period	8/05/14 → 9/05/14

Keywords

auto-regressive model
extended morphological profile
hyperspectral images
random forest

Access to Document

10.1109/IWECA.2014.6845627

Cite this

Ren, Y., Zhang, Y., Wei, W., & Li, L. (2014). A spectral-spatial hyperspectral data classification approach using random forest with label constraints. In Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014 (pp. 344-347). Article 6845627 (Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014). IEEE Computer Society. https://doi.org/10.1109/IWECA.2014.6845627

Ren, Yuemei ; Zhang, Yanning ; Wei, Wei et al. / A spectral-spatial hyperspectral data classification approach using random forest with label constraints. Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014. IEEE Computer Society, 2014. pp. 344-347 (Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014).

@inproceedings{732d981f06ee434786f82f0ad87e45c7,

title = "A spectral-spatial hyperspectral data classification approach using random forest with label constraints",

abstract = "A new classification approach using random forest with label constraints is proposed to deal with the underutilization effectively of spectral and spatial information for hyperspectral image classification. Firstly, the principal component analysis extraction method is adopted, and the extended morphological profiles of the image are extracted from the principle components images by using mathematical morphology method. Then random forest is constructed based on the extracted features. Finally, the label constraints based on space continuity is used to constraint the results by using the label information of its neighborhoods on image space. The classification result is decided by voting strategy. Experimental results of several real hyperspectral images demonstrate that the proposed approach outperforms the random forest method without constraint and the popular SVM classification method.",

keywords = "auto-regressive model, extended morphological profile, hyperspectral images, random forest",

author = "Yuemei Ren and Yanning Zhang and Wei Wei and Lei Li",

year = "2014",

doi = "10.1109/IWECA.2014.6845627",

language = "英语",

isbn = "9781479945658",

series = "Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014",

publisher = "IEEE Computer Society",

pages = "344--347",

booktitle = "Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014",

note = "2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014 ; Conference date: 08-05-2014 Through 09-05-2014",

}

Ren, Y, Zhang, Y , Wei, W & Li, L 2014, A spectral-spatial hyperspectral data classification approach using random forest with label constraints. in Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014., 6845627, Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014, IEEE Computer Society, pp. 344-347, 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014, Ottawa, ON, Canada, 8/05/14. https://doi.org/10.1109/IWECA.2014.6845627

A spectral-spatial hyperspectral data classification approach using random forest with label constraints. / Ren, Yuemei; Zhang, Yanning ; Wei, Wei et al.
Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014. IEEE Computer Society, 2014. p. 344-347 6845627 (Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014).

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

TY - GEN

T1 - A spectral-spatial hyperspectral data classification approach using random forest with label constraints

AU - Ren, Yuemei

AU - Zhang, Yanning

AU - Wei, Wei

AU - Li, Lei

PY - 2014

Y1 - 2014

N2 - A new classification approach using random forest with label constraints is proposed to deal with the underutilization effectively of spectral and spatial information for hyperspectral image classification. Firstly, the principal component analysis extraction method is adopted, and the extended morphological profiles of the image are extracted from the principle components images by using mathematical morphology method. Then random forest is constructed based on the extracted features. Finally, the label constraints based on space continuity is used to constraint the results by using the label information of its neighborhoods on image space. The classification result is decided by voting strategy. Experimental results of several real hyperspectral images demonstrate that the proposed approach outperforms the random forest method without constraint and the popular SVM classification method.

AB - A new classification approach using random forest with label constraints is proposed to deal with the underutilization effectively of spectral and spatial information for hyperspectral image classification. Firstly, the principal component analysis extraction method is adopted, and the extended morphological profiles of the image are extracted from the principle components images by using mathematical morphology method. Then random forest is constructed based on the extracted features. Finally, the label constraints based on space continuity is used to constraint the results by using the label information of its neighborhoods on image space. The classification result is decided by voting strategy. Experimental results of several real hyperspectral images demonstrate that the proposed approach outperforms the random forest method without constraint and the popular SVM classification method.

KW - auto-regressive model

KW - extended morphological profile

KW - hyperspectral images

KW - random forest

UR - https://www.scopus.com/pages/publications/84904540512

U2 - 10.1109/IWECA.2014.6845627

DO - 10.1109/IWECA.2014.6845627

M3 - 会议稿件

AN - SCOPUS:84904540512

SN - 9781479945658

T3 - Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014

SP - 344

EP - 347

BT - Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014

PB - IEEE Computer Society

T2 - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014

Y2 - 8 May 2014 through 9 May 2014

ER -

Ren Y, Zhang Y , Wei W, Li L. A spectral-spatial hyperspectral data classification approach using random forest with label constraints. In Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014. IEEE Computer Society. 2014. p. 344-347. 6845627. (Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014). doi: 10.1109/IWECA.2014.6845627

A spectral-spatial hyperspectral data classification approach using random forest with label constraints

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this