Cost-Sensitive Self-Paced Learning With Adaptive Regularization for Classification of Image Time Series

The classification of image time series has potential significance in the field of land-cover analysis with the increasing number of remote sensing images. The key problem of the classification of image time series is how to transfer the already available knowledge on the source domain to the target...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Hao Li, Jianzhao Li, Yue Zhao, Maoguo Gong, Yujing Zhang, Tongfei Liu
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Classification cost-sensitive self-paced learning time series Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
Acceso en línea:	https://doaj.org/article/714a97b0b02343f59ea68bfbad0a11f7
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:714a97b0b02343f59ea68bfbad0a11f7
record_format	dspace
spelling	oai:doaj.org-article:714a97b0b02343f59ea68bfbad0a11f72021-12-02T00:00:08ZCost-Sensitive Self-Paced Learning With Adaptive Regularization for Classification of Image Time Series2151-153510.1109/JSTARS.2021.3127754https://doaj.org/article/714a97b0b02343f59ea68bfbad0a11f72021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9613766/https://doaj.org/toc/2151-1535The classification of image time series has potential significance in the field of land-cover analysis with the increasing number of remote sensing images. The key problem of the classification of image time series is how to transfer the already available knowledge on the source domain to the target domain. Nevertheless, most of the existing methods do not consider the impact of different sample costs on the classifier during transferring. In addition, it is very difficult to collect reliable labeled samples with changed or unchanged categories between the source domain and the target domain in the case of a large number of training samples. In order to alleviate the above problems, we propose a cost-sensitive self-paced learning (CSSPL) framework with adaptive regularization for the classification of image time series in this article. Considering that the costs of different samples cannot be completely equal to the classifier, different cost values are assigned to each type of error first, then we minimize the total cost to give the change detection classifier a preference on the unchanged class, aiming to reduce wrong label propagation from the source to the target image. Besides, an adaptive mixture weight regularizer is designed to automatically assign sample weight based on the loss value of the dataset, which enables more reliable sample weights to be selected for training. Experimental results show that the proposed algorithm provides a set of reliable samples for the training of classifier and achieves a promising improvement on classification accuracy.Hao LiJianzhao LiYue ZhaoMaoguo GongYujing ZhangTongfei LiuIEEEarticleClassificationcost-sensitiveself-paced learningtime seriesOcean engineeringTC1501-1800Geophysics. Cosmic physicsQC801-809ENIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 14, Pp 11713-11727 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Classification cost-sensitive self-paced learning time series Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
spellingShingle	Classification cost-sensitive self-paced learning time series Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809 Hao Li Jianzhao Li Yue Zhao Maoguo Gong Yujing Zhang Tongfei Liu Cost-Sensitive Self-Paced Learning With Adaptive Regularization for Classification of Image Time Series
description	The classification of image time series has potential significance in the field of land-cover analysis with the increasing number of remote sensing images. The key problem of the classification of image time series is how to transfer the already available knowledge on the source domain to the target domain. Nevertheless, most of the existing methods do not consider the impact of different sample costs on the classifier during transferring. In addition, it is very difficult to collect reliable labeled samples with changed or unchanged categories between the source domain and the target domain in the case of a large number of training samples. In order to alleviate the above problems, we propose a cost-sensitive self-paced learning (CSSPL) framework with adaptive regularization for the classification of image time series in this article. Considering that the costs of different samples cannot be completely equal to the classifier, different cost values are assigned to each type of error first, then we minimize the total cost to give the change detection classifier a preference on the unchanged class, aiming to reduce wrong label propagation from the source to the target image. Besides, an adaptive mixture weight regularizer is designed to automatically assign sample weight based on the loss value of the dataset, which enables more reliable sample weights to be selected for training. Experimental results show that the proposed algorithm provides a set of reliable samples for the training of classifier and achieves a promising improvement on classification accuracy.
format	article
author	Hao Li Jianzhao Li Yue Zhao Maoguo Gong Yujing Zhang Tongfei Liu
author_facet	Hao Li Jianzhao Li Yue Zhao Maoguo Gong Yujing Zhang Tongfei Liu
author_sort	Hao Li
title	Cost-Sensitive Self-Paced Learning With Adaptive Regularization for Classification of Image Time Series
title_short	Cost-Sensitive Self-Paced Learning With Adaptive Regularization for Classification of Image Time Series
title_full	Cost-Sensitive Self-Paced Learning With Adaptive Regularization for Classification of Image Time Series
title_fullStr	Cost-Sensitive Self-Paced Learning With Adaptive Regularization for Classification of Image Time Series
title_full_unstemmed	Cost-Sensitive Self-Paced Learning With Adaptive Regularization for Classification of Image Time Series
title_sort	cost-sensitive self-paced learning with adaptive regularization for classification of image time series
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/714a97b0b02343f59ea68bfbad0a11f7
work_keys_str_mv	AT haoli costsensitiveselfpacedlearningwithadaptiveregularizationforclassificationofimagetimeseries AT jianzhaoli costsensitiveselfpacedlearningwithadaptiveregularizationforclassificationofimagetimeseries AT yuezhao costsensitiveselfpacedlearningwithadaptiveregularizationforclassificationofimagetimeseries AT maoguogong costsensitiveselfpacedlearningwithadaptiveregularizationforclassificationofimagetimeseries AT yujingzhang costsensitiveselfpacedlearningwithadaptiveregularizationforclassificationofimagetimeseries AT tongfeiliu costsensitiveselfpacedlearningwithadaptiveregularizationforclassificationofimagetimeseries
_version_	1718403991579131904

Cost-Sensitive Self-Paced Learning With Adaptive Regularization for Classification of Image Time Series

Ejemplares similares