Aurora Image Classification Based on Multi-Feature Latent Dirichlet Allocation
Due to the rich physical meaning of aurora morphology, the classification of aurora images is an important task for polar scientific expeditions. However, the traditional classification methods do not make full use of the different features of aurora images, and the dimension of the description feat...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2018
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs10020233 |
id |
ftmdpi:oai:mdpi.com:/2072-4292/10/2/233/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/2072-4292/10/2/233/ 2023-08-20T04:09:23+02:00 Aurora Image Classification Based on Multi-Feature Latent Dirichlet Allocation Yanfei Zhong Rui Huang Ji Zhao Bei Zhao Tingting Liu agris 2018-02-03 application/pdf https://doi.org/10.3390/rs10020233 EN eng Multidisciplinary Digital Publishing Institute Remote Sensing Image Processing https://dx.doi.org/10.3390/rs10020233 https://creativecommons.org/licenses/by/4.0/ Remote Sensing; Volume 10; Issue 2; Pages: 233 aurora image classification multiple features 1-D histogram latent Dirichlet allocation (LDA) probabilistic topic model (PTM) Text 2018 ftmdpi https://doi.org/10.3390/rs10020233 2023-07-31T21:22:37Z Due to the rich physical meaning of aurora morphology, the classification of aurora images is an important task for polar scientific expeditions. However, the traditional classification methods do not make full use of the different features of aurora images, and the dimension of the description features is usually so high that it reduces the efficiency. In this paper, through combining multiple features extracted from aurora images, an aurora image classification method based on multi-feature latent Dirichlet allocation (AI-MFLDA) is proposed. Different types of features, whether local or global, discrete or continuous, can be integrated after being transformed to one-dimensional (1-D) histograms, and the dimension of the description features can be reduced due to using only a few topics to represent the aurora images. In the experiments, according to the classification system provided by the Polar Research Institute of China, a four-class aurora image dataset was tested and three types of features (MeanStd, scale-invariant feature transform (SIFT), and shape-based invariant texture index (SITI)) were utilized. The experimental results showed that, compared to the traditional methods, the proposed AI-MFLDA is able to achieve a better performance with 98.2% average classification accuracy while maintaining a low feature dimension. Text Polar Research Institute of China MDPI Open Access Publishing Remote Sensing 10 2 233 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
aurora image classification multiple features 1-D histogram latent Dirichlet allocation (LDA) probabilistic topic model (PTM) |
spellingShingle |
aurora image classification multiple features 1-D histogram latent Dirichlet allocation (LDA) probabilistic topic model (PTM) Yanfei Zhong Rui Huang Ji Zhao Bei Zhao Tingting Liu Aurora Image Classification Based on Multi-Feature Latent Dirichlet Allocation |
topic_facet |
aurora image classification multiple features 1-D histogram latent Dirichlet allocation (LDA) probabilistic topic model (PTM) |
description |
Due to the rich physical meaning of aurora morphology, the classification of aurora images is an important task for polar scientific expeditions. However, the traditional classification methods do not make full use of the different features of aurora images, and the dimension of the description features is usually so high that it reduces the efficiency. In this paper, through combining multiple features extracted from aurora images, an aurora image classification method based on multi-feature latent Dirichlet allocation (AI-MFLDA) is proposed. Different types of features, whether local or global, discrete or continuous, can be integrated after being transformed to one-dimensional (1-D) histograms, and the dimension of the description features can be reduced due to using only a few topics to represent the aurora images. In the experiments, according to the classification system provided by the Polar Research Institute of China, a four-class aurora image dataset was tested and three types of features (MeanStd, scale-invariant feature transform (SIFT), and shape-based invariant texture index (SITI)) were utilized. The experimental results showed that, compared to the traditional methods, the proposed AI-MFLDA is able to achieve a better performance with 98.2% average classification accuracy while maintaining a low feature dimension. |
format |
Text |
author |
Yanfei Zhong Rui Huang Ji Zhao Bei Zhao Tingting Liu |
author_facet |
Yanfei Zhong Rui Huang Ji Zhao Bei Zhao Tingting Liu |
author_sort |
Yanfei Zhong |
title |
Aurora Image Classification Based on Multi-Feature Latent Dirichlet Allocation |
title_short |
Aurora Image Classification Based on Multi-Feature Latent Dirichlet Allocation |
title_full |
Aurora Image Classification Based on Multi-Feature Latent Dirichlet Allocation |
title_fullStr |
Aurora Image Classification Based on Multi-Feature Latent Dirichlet Allocation |
title_full_unstemmed |
Aurora Image Classification Based on Multi-Feature Latent Dirichlet Allocation |
title_sort |
aurora image classification based on multi-feature latent dirichlet allocation |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2018 |
url |
https://doi.org/10.3390/rs10020233 |
op_coverage |
agris |
genre |
Polar Research Institute of China |
genre_facet |
Polar Research Institute of China |
op_source |
Remote Sensing; Volume 10; Issue 2; Pages: 233 |
op_relation |
Remote Sensing Image Processing https://dx.doi.org/10.3390/rs10020233 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/rs10020233 |
container_title |
Remote Sensing |
container_volume |
10 |
container_issue |
2 |
container_start_page |
233 |
_version_ |
1774722279918272512 |