Multitemporal ensemble learning for snow cover extraction from high-spatial-resolution images in mountain areas

Pengfeng Xiao, Chengxi Li, Liujun Zhu, Xueliang Zhang, Tengyao Ma, Xuezhi Feng

Research output: Contribution to journalArticleResearchpeer-review

Abstract

High-spatial and -temporal resolution snow cover products in mountain areas are important to hydrological applications. The GF-1 satellite provides multispectral images with 8-m resolution and a revisit up to 2 days, which makes it possible to produce snow cover products. However, it is challenging to extract snow cover from these images because of limited spectral bands, severe mountain shadows, and dataset-shift problem in multitemporal classification. To overcome the limitations above, this study proposes a multitemporal ensemble learning framework to extract snow cover from high-spatial-resolution images in mountain areas. The principle behind ensemble learning, i.e. learning from disagreement, is extended from single image classification to multitemporal ones. We assume that multitemporal training samples selected within time-invariant classes at the same locations can be different in feature space. Such disagreements are used in multitemporal ensemble learning to improve classification accuracy. To enhance both accuracy and diversity of the multiple classifiers trained on these samples, a joint feature selection method is suggested to select the optimal multitemporal feature space and a joint parameter optimization method is designed to ensemble classifiers trained for multitemporal images. The experiments show that the performances of multitemporal ensemble classifiers are superior to that of single classifiers, confirming the effectiveness of the proposed framework.

Original languageEnglish
Pages (from-to)1668-1691
Number of pages24
JournalInternational Journal of Remote Sensing
Volume41
Issue number5
DOIs
Publication statusPublished - 2020

Cite this

Xiao, Pengfeng ; Li, Chengxi ; Zhu, Liujun ; Zhang, Xueliang ; Ma, Tengyao ; Feng, Xuezhi. / Multitemporal ensemble learning for snow cover extraction from high-spatial-resolution images in mountain areas. In: International Journal of Remote Sensing. 2020 ; Vol. 41, No. 5. pp. 1668-1691.
@article{fad019d56ec7473aa97b52352f597d30,
title = "Multitemporal ensemble learning for snow cover extraction from high-spatial-resolution images in mountain areas",
abstract = "High-spatial and -temporal resolution snow cover products in mountain areas are important to hydrological applications. The GF-1 satellite provides multispectral images with 8-m resolution and a revisit up to 2 days, which makes it possible to produce snow cover products. However, it is challenging to extract snow cover from these images because of limited spectral bands, severe mountain shadows, and dataset-shift problem in multitemporal classification. To overcome the limitations above, this study proposes a multitemporal ensemble learning framework to extract snow cover from high-spatial-resolution images in mountain areas. The principle behind ensemble learning, i.e. learning from disagreement, is extended from single image classification to multitemporal ones. We assume that multitemporal training samples selected within time-invariant classes at the same locations can be different in feature space. Such disagreements are used in multitemporal ensemble learning to improve classification accuracy. To enhance both accuracy and diversity of the multiple classifiers trained on these samples, a joint feature selection method is suggested to select the optimal multitemporal feature space and a joint parameter optimization method is designed to ensemble classifiers trained for multitemporal images. The experiments show that the performances of multitemporal ensemble classifiers are superior to that of single classifiers, confirming the effectiveness of the proposed framework.",
author = "Pengfeng Xiao and Chengxi Li and Liujun Zhu and Xueliang Zhang and Tengyao Ma and Xuezhi Feng",
year = "2020",
doi = "10.1080/01431161.2019.1674458",
language = "English",
volume = "41",
pages = "1668--1691",
journal = "International Journal of Remote Sensing",
issn = "0143-1161",
publisher = "Taylor & Francis",
number = "5",

}

Multitemporal ensemble learning for snow cover extraction from high-spatial-resolution images in mountain areas. / Xiao, Pengfeng; Li, Chengxi; Zhu, Liujun; Zhang, Xueliang; Ma, Tengyao; Feng, Xuezhi.

In: International Journal of Remote Sensing, Vol. 41, No. 5, 2020, p. 1668-1691.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Multitemporal ensemble learning for snow cover extraction from high-spatial-resolution images in mountain areas

AU - Xiao, Pengfeng

AU - Li, Chengxi

AU - Zhu, Liujun

AU - Zhang, Xueliang

AU - Ma, Tengyao

AU - Feng, Xuezhi

PY - 2020

Y1 - 2020

N2 - High-spatial and -temporal resolution snow cover products in mountain areas are important to hydrological applications. The GF-1 satellite provides multispectral images with 8-m resolution and a revisit up to 2 days, which makes it possible to produce snow cover products. However, it is challenging to extract snow cover from these images because of limited spectral bands, severe mountain shadows, and dataset-shift problem in multitemporal classification. To overcome the limitations above, this study proposes a multitemporal ensemble learning framework to extract snow cover from high-spatial-resolution images in mountain areas. The principle behind ensemble learning, i.e. learning from disagreement, is extended from single image classification to multitemporal ones. We assume that multitemporal training samples selected within time-invariant classes at the same locations can be different in feature space. Such disagreements are used in multitemporal ensemble learning to improve classification accuracy. To enhance both accuracy and diversity of the multiple classifiers trained on these samples, a joint feature selection method is suggested to select the optimal multitemporal feature space and a joint parameter optimization method is designed to ensemble classifiers trained for multitemporal images. The experiments show that the performances of multitemporal ensemble classifiers are superior to that of single classifiers, confirming the effectiveness of the proposed framework.

AB - High-spatial and -temporal resolution snow cover products in mountain areas are important to hydrological applications. The GF-1 satellite provides multispectral images with 8-m resolution and a revisit up to 2 days, which makes it possible to produce snow cover products. However, it is challenging to extract snow cover from these images because of limited spectral bands, severe mountain shadows, and dataset-shift problem in multitemporal classification. To overcome the limitations above, this study proposes a multitemporal ensemble learning framework to extract snow cover from high-spatial-resolution images in mountain areas. The principle behind ensemble learning, i.e. learning from disagreement, is extended from single image classification to multitemporal ones. We assume that multitemporal training samples selected within time-invariant classes at the same locations can be different in feature space. Such disagreements are used in multitemporal ensemble learning to improve classification accuracy. To enhance both accuracy and diversity of the multiple classifiers trained on these samples, a joint feature selection method is suggested to select the optimal multitemporal feature space and a joint parameter optimization method is designed to ensemble classifiers trained for multitemporal images. The experiments show that the performances of multitemporal ensemble classifiers are superior to that of single classifiers, confirming the effectiveness of the proposed framework.

UR - http://www.scopus.com/inward/record.url?scp=85074026362&partnerID=8YFLogxK

U2 - 10.1080/01431161.2019.1674458

DO - 10.1080/01431161.2019.1674458

M3 - Article

AN - SCOPUS:85074026362

VL - 41

SP - 1668

EP - 1691

JO - International Journal of Remote Sensing

JF - International Journal of Remote Sensing

SN - 0143-1161

IS - 5

ER -