Cross-Domain Ground-Based Cloud Classification Based on Transfer of Local Features and Discriminative Metric Learning
Cross-domain ground-based cloud classification is a challenging issue as the appearance of cloud images from different cloud databases possesses extreme variations. Two fundamental problems which are essential for cross-domain ground-based cloud classification are feature representation and similari...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
MDPI AG
2017
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs10010008 https://doaj.org/article/da97b0b61bbb4e93a0bfc43006a71341 |
id |
ftdoajarticles:oai:doaj.org/article:da97b0b61bbb4e93a0bfc43006a71341 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:da97b0b61bbb4e93a0bfc43006a71341 2023-05-15T16:01:44+02:00 Cross-Domain Ground-Based Cloud Classification Based on Transfer of Local Features and Discriminative Metric Learning Zhong Zhang Donghong Li Shuang Liu Baihua Xiao Xiaozhong Cao 2017-12-01T00:00:00Z https://doi.org/10.3390/rs10010008 https://doaj.org/article/da97b0b61bbb4e93a0bfc43006a71341 EN eng MDPI AG https://www.mdpi.com/2072-4292/10/1/8 https://doaj.org/toc/2072-4292 2072-4292 doi:10.3390/rs10010008 https://doaj.org/article/da97b0b61bbb4e93a0bfc43006a71341 Remote Sensing, Vol 10, Iss 1, p 8 (2017) ground-based cloud classification machine learning transfer of local features discriminative metric learning Science Q article 2017 ftdoajarticles https://doi.org/10.3390/rs10010008 2022-12-31T15:16:41Z Cross-domain ground-based cloud classification is a challenging issue as the appearance of cloud images from different cloud databases possesses extreme variations. Two fundamental problems which are essential for cross-domain ground-based cloud classification are feature representation and similarity measurement. In this paper, we propose an effective feature representation called transfer of local features (TLF), and measurement method called discriminative metric learning (DML). The TLF is a generalized representation framework that can integrate various kinds of local features, e.g., local binary patterns (LBP), local ternary patterns (LTP) and completed LBP (CLBP). In order to handle domain shift, such as variations of illumination, image resolution, capturing location, occlusion and so on, the TLF mines the maximum response in regions to make a stable representation for domain variations. We also propose to learn a discriminant metric, simultaneously. We make use of sample pairs and the relationship among cloud classes to learn the distance metric. Furthermore, in order to improve the practicability of the proposed method, we replace the original cloud images with the convolutional activation maps which are then applied to TLF and DML. The proposed method has been validated on three cloud databases which are collected in China alone, provided by Chinese Academy of Meteorological Sciences (CAMS), Meteorological Observation Centre (MOC), and Institute of Atmospheric Physics (IAP). The classification accuracies outperform the state-of-the-art methods. Article in Journal/Newspaper DML Directory of Open Access Journals: DOAJ Articles Remote Sensing 10 2 8 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
ground-based cloud classification machine learning transfer of local features discriminative metric learning Science Q |
spellingShingle |
ground-based cloud classification machine learning transfer of local features discriminative metric learning Science Q Zhong Zhang Donghong Li Shuang Liu Baihua Xiao Xiaozhong Cao Cross-Domain Ground-Based Cloud Classification Based on Transfer of Local Features and Discriminative Metric Learning |
topic_facet |
ground-based cloud classification machine learning transfer of local features discriminative metric learning Science Q |
description |
Cross-domain ground-based cloud classification is a challenging issue as the appearance of cloud images from different cloud databases possesses extreme variations. Two fundamental problems which are essential for cross-domain ground-based cloud classification are feature representation and similarity measurement. In this paper, we propose an effective feature representation called transfer of local features (TLF), and measurement method called discriminative metric learning (DML). The TLF is a generalized representation framework that can integrate various kinds of local features, e.g., local binary patterns (LBP), local ternary patterns (LTP) and completed LBP (CLBP). In order to handle domain shift, such as variations of illumination, image resolution, capturing location, occlusion and so on, the TLF mines the maximum response in regions to make a stable representation for domain variations. We also propose to learn a discriminant metric, simultaneously. We make use of sample pairs and the relationship among cloud classes to learn the distance metric. Furthermore, in order to improve the practicability of the proposed method, we replace the original cloud images with the convolutional activation maps which are then applied to TLF and DML. The proposed method has been validated on three cloud databases which are collected in China alone, provided by Chinese Academy of Meteorological Sciences (CAMS), Meteorological Observation Centre (MOC), and Institute of Atmospheric Physics (IAP). The classification accuracies outperform the state-of-the-art methods. |
format |
Article in Journal/Newspaper |
author |
Zhong Zhang Donghong Li Shuang Liu Baihua Xiao Xiaozhong Cao |
author_facet |
Zhong Zhang Donghong Li Shuang Liu Baihua Xiao Xiaozhong Cao |
author_sort |
Zhong Zhang |
title |
Cross-Domain Ground-Based Cloud Classification Based on Transfer of Local Features and Discriminative Metric Learning |
title_short |
Cross-Domain Ground-Based Cloud Classification Based on Transfer of Local Features and Discriminative Metric Learning |
title_full |
Cross-Domain Ground-Based Cloud Classification Based on Transfer of Local Features and Discriminative Metric Learning |
title_fullStr |
Cross-Domain Ground-Based Cloud Classification Based on Transfer of Local Features and Discriminative Metric Learning |
title_full_unstemmed |
Cross-Domain Ground-Based Cloud Classification Based on Transfer of Local Features and Discriminative Metric Learning |
title_sort |
cross-domain ground-based cloud classification based on transfer of local features and discriminative metric learning |
publisher |
MDPI AG |
publishDate |
2017 |
url |
https://doi.org/10.3390/rs10010008 https://doaj.org/article/da97b0b61bbb4e93a0bfc43006a71341 |
genre |
DML |
genre_facet |
DML |
op_source |
Remote Sensing, Vol 10, Iss 1, p 8 (2017) |
op_relation |
https://www.mdpi.com/2072-4292/10/1/8 https://doaj.org/toc/2072-4292 2072-4292 doi:10.3390/rs10010008 https://doaj.org/article/da97b0b61bbb4e93a0bfc43006a71341 |
op_doi |
https://doi.org/10.3390/rs10010008 |
container_title |
Remote Sensing |
container_volume |
10 |
container_issue |
2 |
container_start_page |
8 |
_version_ |
1766397479423574016 |