Novel semi-supervised classification method based on class certainty of samples

Fei Gao, Zhenyu Yue, Jun Wang, Erfu Yang, Amir Hussain

Research output: Contribution to conferencePaper

Abstract

The traditional classification method based on supervised learning classifies remote sensing (RS) images by using sufficient labeled samples. However, the number of labeled samples is limited due to the expensive and time-consuming collection. To effectively utilize the information of unlabeled 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 unlabeled 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. Ultimately, we extend the semi-supervised LDA to nonlinear dimensional reduction by combining class certainty and kernel methods. Furthermore, to assess the effectiveness of proposed method, the nearest neighbor classifier is adopted to classify actual SAR images. The results demonstrate that the proposed method can effectively exploit the information of unlabeled samples and greatly improve the classification effect compared with other state-of-the-art approaches.

Conference

ConferenceThe 9th International Conference on Brain-Inspired Cognitive System​
Abbreviated titleBICS2018
CountryChina
CityXi'an
Period7/07/188/07/18
Internet address

Fingerprint

Discriminant analysis
Supervised learning
Remote sensing
Classifiers

Keywords

  • remote sensing images
  • semi-supervised classification
  • class certainty
  • semi-supervised LDA
  • kernal method

Cite this

Gao, F., Yue, Z., Wang, J., Yang, E., & Hussain, A. (2018). Novel semi-supervised classification method based on class certainty of samples. 1-10. Paper presented at The 9th International Conference on Brain-Inspired Cognitive System​, Xi'an , China.
Gao, Fei ; Yue, Zhenyu ; Wang, Jun ; Yang, Erfu ; Hussain, Amir. / Novel semi-supervised classification method based on class certainty of samples. Paper presented at The 9th International Conference on Brain-Inspired Cognitive System​, Xi'an , China.10 p.
@conference{e1794f494b404dd38be20b350a17f14f,
title = "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 labeled samples. However, the number of labeled samples is limited due to the expensive and time-consuming collection. To effectively utilize the information of unlabeled 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 unlabeled 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. Ultimately, we extend the semi-supervised LDA to nonlinear dimensional reduction by combining class certainty and kernel methods. Furthermore, to assess the effectiveness of proposed method, the nearest neighbor classifier is adopted to classify actual SAR images. The results demonstrate that the proposed method can effectively exploit the information of unlabeled samples and greatly improve the classification effect compared with other state-of-the-art approaches.",
keywords = "remote sensing images, semi-supervised classification, class certainty, semi-supervised LDA, kernal method",
author = "Fei Gao and Zhenyu Yue and Jun Wang and Erfu Yang and Amir Hussain",
year = "2018",
month = "7",
day = "7",
language = "English",
pages = "1--10",
note = "The 9th International Conference on Brain-Inspired Cognitive System​, BICS2018 ; Conference date: 07-07-2018 Through 08-07-2018",
url = "http://bics2018.org/",

}

Gao, F, Yue, Z, Wang, J, Yang, E & Hussain, A 2018, 'Novel semi-supervised classification method based on class certainty of samples' Paper presented at The 9th International Conference on Brain-Inspired Cognitive System​, Xi'an , China, 7/07/18 - 8/07/18, pp. 1-10.

Novel semi-supervised classification method based on class certainty of samples. / Gao, Fei; Yue, Zhenyu; Wang, Jun; Yang, Erfu; Hussain, Amir.

2018. 1-10 Paper presented at The 9th International Conference on Brain-Inspired Cognitive System​, Xi'an , China.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Novel semi-supervised classification method based on class certainty of samples

AU - Gao, Fei

AU - Yue, Zhenyu

AU - Wang, Jun

AU - Yang, Erfu

AU - Hussain, Amir

PY - 2018/7/7

Y1 - 2018/7/7

N2 - The traditional classification method based on supervised learning classifies remote sensing (RS) images by using sufficient labeled samples. However, the number of labeled samples is limited due to the expensive and time-consuming collection. To effectively utilize the information of unlabeled 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 unlabeled 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. Ultimately, we extend the semi-supervised LDA to nonlinear dimensional reduction by combining class certainty and kernel methods. Furthermore, to assess the effectiveness of proposed method, the nearest neighbor classifier is adopted to classify actual SAR images. The results demonstrate that the proposed method can effectively exploit the information of unlabeled 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 labeled samples. However, the number of labeled samples is limited due to the expensive and time-consuming collection. To effectively utilize the information of unlabeled 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 unlabeled 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. Ultimately, we extend the semi-supervised LDA to nonlinear dimensional reduction by combining class certainty and kernel methods. Furthermore, to assess the effectiveness of proposed method, the nearest neighbor classifier is adopted to classify actual SAR images. The results demonstrate that the proposed method can effectively exploit the information of unlabeled samples and greatly improve the classification effect compared with other state-of-the-art approaches.

KW - remote sensing images

KW - semi-supervised classification

KW - class certainty

KW - semi-supervised LDA

KW - kernal method

M3 - Paper

SP - 1

EP - 10

ER -

Gao F, Yue Z, Wang J, Yang E, Hussain A. Novel semi-supervised classification method based on class certainty of samples. 2018. Paper presented at The 9th International Conference on Brain-Inspired Cognitive System​, Xi'an , China.