A Novel Semi-supervised Classification Method Based on Class Certainty of Samples

Fei Gao; Zhenyu Yue; Qingxu Xiong; Jun Wang; Erfu Yang; Amir Hussain

doi:10.1007/978-3-030-00563-4_30

A Novel Semi-supervised Classification Method Based on Class Certainty of Samples

Fei Gao
, Zhenyu Yue^*
, Qingxu Xiong
, Jun Wang
, Erfu Yang
, Amir Hussain

^*Corresponding author for this work

School of Electronic and Information Engineering

Beihang University
University of Strathclyde
University of Stirling

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

1 Scopus citations

Abstract

The traditional classification method based on supervised learning classifies remote sensing (RS) images by using sufficient labelled samples. However, the number of labelled samples is limited due to the expensive and time-consuming collection. To effectively utilize the information of unlabelled samples in the learning process, this paper proposes a novel semi-supervised classification method based on class certainty of samples (CCS). First, the class certainty of unlabelled samples obtained based on multi-class SVM is smoothed for robustness. Then, a new semi-supervised linear discriminant analysis (LDA) is presented based on class certainty, which improves the separability of samples in the projection subspace. Finally, the nearest neighbor classifier is adopted to classify the images. The experimental results demonstrate that the proposed method can effectively exploit the information of unlabelled samples and greatly improve the classification effect compared with other state-of-the-art approaches.

Original language	English
Title of host publication	Advances in Brain Inspired Cognitive Systems - 9th International Conference, BICS 2018, Proceedings
Editors	Amir Hussain, Bin Luo, Jiangbin Zheng, Xinbo Zhao, Cheng-Lin Liu, Jinchang Ren, Huimin Zhao
Publisher	Springer Verlag
Pages	315-324
Number of pages	10
ISBN (Print)	9783030005627
DOIs	https://doi.org/10.1007/978-3-030-00563-4_30
State	Published - 2018
Event	9th International Conference on Brain-Inspired Cognitive Systems, BICS 2018 - Xi'an, China Duration: 7 Jul 2018 → 8 Jul 2018

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10989 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	9th International Conference on Brain-Inspired Cognitive Systems, BICS 2018
Country/Territory	China
City	Xi'an
Period	7/07/18 → 8/07/18

Keywords

Class certainty
Remote sensing images
Semi-supervised LDA
Semi-supervised classification

Access to Document

10.1007/978-3-030-00563-4_30

Cite this

Gao, F., Yue, Z., Xiong, Q., Wang, J., Yang, E., & Hussain, A. (2018). A Novel Semi-supervised Classification Method Based on Class Certainty of Samples. In A. Hussain, B. Luo, J. Zheng, X. Zhao, C.-L. Liu, J. Ren, & H. Zhao (Eds.), Advances in Brain Inspired Cognitive Systems - 9th International Conference, BICS 2018, Proceedings (pp. 315-324). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10989 LNAI). Springer Verlag. https://doi.org/10.1007/978-3-030-00563-4_30

Gao, Fei ; Yue, Zhenyu ; Xiong, Qingxu et al. / A Novel Semi-supervised Classification Method Based on Class Certainty of Samples. Advances in Brain Inspired Cognitive Systems - 9th International Conference, BICS 2018, Proceedings. editor / Amir Hussain ; Bin Luo ; Jiangbin Zheng ; Xinbo Zhao ; Cheng-Lin Liu ; Jinchang Ren ; Huimin Zhao. Springer Verlag, 2018. pp. 315-324 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{e857f6d0837a4befa5027d564911d80f,

title = "A Novel Semi-supervised Classification Method Based on Class Certainty of Samples",

abstract = "The traditional classification method based on supervised learning classifies remote sensing (RS) images by using sufficient labelled samples. However, the number of labelled samples is limited due to the expensive and time-consuming collection. To effectively utilize the information of unlabelled samples in the learning process, this paper proposes a novel semi-supervised classification method based on class certainty of samples (CCS). First, the class certainty of unlabelled samples obtained based on multi-class SVM is smoothed for robustness. Then, a new semi-supervised linear discriminant analysis (LDA) is presented based on class certainty, which improves the separability of samples in the projection subspace. Finally, the nearest neighbor classifier is adopted to classify the images. The experimental results demonstrate that the proposed method can effectively exploit the information of unlabelled samples and greatly improve the classification effect compared with other state-of-the-art approaches.",

keywords = "Class certainty, Remote sensing images, Semi-supervised LDA, Semi-supervised classification",

author = "Fei Gao and Zhenyu Yue and Qingxu Xiong and Jun Wang and Erfu Yang and Amir Hussain",

note = "Publisher Copyright: {\textcopyright} 2018, Springer Nature Switzerland AG.; 9th International Conference on Brain-Inspired Cognitive Systems, BICS 2018 ; Conference date: 07-07-2018 Through 08-07-2018",

year = "2018",

doi = "10.1007/978-3-030-00563-4\_30",

language = "英语",

isbn = "9783030005627",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "315--324",

editor = "Amir Hussain and Bin Luo and Jiangbin Zheng and Xinbo Zhao and Cheng-Lin Liu and Jinchang Ren and Huimin Zhao",

booktitle = "Advances in Brain Inspired Cognitive Systems - 9th International Conference, BICS 2018, Proceedings",

address = "德国",

}

Gao, F, Yue, Z, Xiong, Q , Wang, J, Yang, E & Hussain, A 2018, A Novel Semi-supervised Classification Method Based on Class Certainty of Samples. in A Hussain, B Luo, J Zheng, X Zhao, C-L Liu, J Ren & H Zhao (eds), Advances in Brain Inspired Cognitive Systems - 9th International Conference, BICS 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10989 LNAI, Springer Verlag, pp. 315-324, 9th International Conference on Brain-Inspired Cognitive Systems, BICS 2018, Xi'an, China, 7/07/18. https://doi.org/10.1007/978-3-030-00563-4_30

A Novel Semi-supervised Classification Method Based on Class Certainty of Samples. / Gao, Fei; Yue, Zhenyu; Xiong, Qingxu et al.
Advances in Brain Inspired Cognitive Systems - 9th International Conference, BICS 2018, Proceedings. ed. / Amir Hussain; Bin Luo; Jiangbin Zheng; Xinbo Zhao; Cheng-Lin Liu; Jinchang Ren; Huimin Zhao. Springer Verlag, 2018. p. 315-324 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10989 LNAI).

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

TY - GEN

T1 - A Novel Semi-supervised Classification Method Based on Class Certainty of Samples

AU - Gao, Fei

AU - Yue, Zhenyu

AU - Xiong, Qingxu

AU - Wang, Jun

AU - Yang, Erfu

AU - Hussain, Amir

PY - 2018

Y1 - 2018

N2 - The traditional classification method based on supervised learning classifies remote sensing (RS) images by using sufficient labelled samples. However, the number of labelled samples is limited due to the expensive and time-consuming collection. To effectively utilize the information of unlabelled samples in the learning process, this paper proposes a novel semi-supervised classification method based on class certainty of samples (CCS). First, the class certainty of unlabelled samples obtained based on multi-class SVM is smoothed for robustness. Then, a new semi-supervised linear discriminant analysis (LDA) is presented based on class certainty, which improves the separability of samples in the projection subspace. Finally, the nearest neighbor classifier is adopted to classify the images. The experimental results demonstrate that the proposed method can effectively exploit the information of unlabelled samples and greatly improve the classification effect compared with other state-of-the-art approaches.

AB - The traditional classification method based on supervised learning classifies remote sensing (RS) images by using sufficient labelled samples. However, the number of labelled samples is limited due to the expensive and time-consuming collection. To effectively utilize the information of unlabelled samples in the learning process, this paper proposes a novel semi-supervised classification method based on class certainty of samples (CCS). First, the class certainty of unlabelled samples obtained based on multi-class SVM is smoothed for robustness. Then, a new semi-supervised linear discriminant analysis (LDA) is presented based on class certainty, which improves the separability of samples in the projection subspace. Finally, the nearest neighbor classifier is adopted to classify the images. The experimental results demonstrate that the proposed method can effectively exploit the information of unlabelled samples and greatly improve the classification effect compared with other state-of-the-art approaches.

KW - Class certainty

KW - Remote sensing images

KW - Semi-supervised LDA

KW - Semi-supervised classification

UR - https://www.scopus.com/pages/publications/85055135427

U2 - 10.1007/978-3-030-00563-4_30

DO - 10.1007/978-3-030-00563-4_30

M3 - 会议稿件

AN - SCOPUS:85055135427

SN - 9783030005627

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 315

EP - 324

BT - Advances in Brain Inspired Cognitive Systems - 9th International Conference, BICS 2018, Proceedings

A2 - Hussain, Amir

A2 - Luo, Bin

A2 - Zheng, Jiangbin

A2 - Zhao, Xinbo

A2 - Liu, Cheng-Lin

A2 - Ren, Jinchang

A2 - Zhao, Huimin

PB - Springer Verlag

T2 - 9th International Conference on Brain-Inspired Cognitive Systems, BICS 2018

Y2 - 7 July 2018 through 8 July 2018

ER -

Gao F, Yue Z, Xiong Q , Wang J, Yang E, Hussain A. A Novel Semi-supervised Classification Method Based on Class Certainty of Samples. In Hussain A, Luo B, Zheng J, Zhao X, Liu CL, Ren J, Zhao H, editors, Advances in Brain Inspired Cognitive Systems - 9th International Conference, BICS 2018, Proceedings. Springer Verlag. 2018. p. 315-324. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-00563-4_30

A Novel Semi-supervised Classification Method Based on Class Certainty of Samples

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this