Joint superpixel and Transformer for high resolution remote sensing image classification
© 2024. The Author(s)..
Deep neural networks combined with superpixel segmentation have proven to be superior to high-resolution remote sensing image (HRI) classification. Currently, most HRI classification methods that combine deep learning and superpixel segmentation use stacking on multiple scales to extract contextual information from segmented objects. However, this approach does not take into account the contextual dependencies between each segmented object. To solve this problem, a joint superpixel and Transformer (JST) framework is proposed for HRI classification. In JST, HRI is first segmented into superpixel objects as input, and Transformer is used to model the long-range dependencies. The contextual relationship between each input superpixel object is obtained and the class of analyzed objects is output by designing an encoding and decoding Transformer. Additionally, we explore the effect of semantic range on classification accuracy. JST is also tested by using two HRI datasets with overall classification accuracy, average accuracy and Kappa coefficients of 0.79, 0.70, 0.78 and 0.91, 0.85, 0.89, respectively. The effectiveness of the proposed method is compared qualitatively and quantitatively, and the results achieve competitive and consistently better than the benchmark comparison method.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2024 |
---|---|
Erschienen: |
2024 |
Enthalten in: |
Zur Gesamtaufnahme - volume:14 |
---|---|
Enthalten in: |
Scientific reports - 14(2024), 1 vom: 01. März, Seite 5054 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Dang, Guangpu [VerfasserIn] |
---|
Links: |
---|
Themen: |
Deep learning |
---|
Anmerkungen: |
Date Revised 03.03.2024 published: Electronic Citation Status PubMed-not-MEDLINE |
---|
doi: |
10.1038/s41598-024-55482-y |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM369140311 |
---|
LEADER | 01000caa a22002652 4500 | ||
---|---|---|---|
001 | NLM369140311 | ||
003 | DE-627 | ||
005 | 20240303232017.0 | ||
007 | cr uuu---uuuuu | ||
008 | 240301s2024 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1038/s41598-024-55482-y |2 doi | |
028 | 5 | 2 | |a pubmed24n1315.xml |
035 | |a (DE-627)NLM369140311 | ||
035 | |a (NLM)38424135 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Dang, Guangpu |e verfasserin |4 aut | |
245 | 1 | 0 | |a Joint superpixel and Transformer for high resolution remote sensing image classification |
264 | 1 | |c 2024 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Revised 03.03.2024 | ||
500 | |a published: Electronic | ||
500 | |a Citation Status PubMed-not-MEDLINE | ||
520 | |a © 2024. The Author(s). | ||
520 | |a Deep neural networks combined with superpixel segmentation have proven to be superior to high-resolution remote sensing image (HRI) classification. Currently, most HRI classification methods that combine deep learning and superpixel segmentation use stacking on multiple scales to extract contextual information from segmented objects. However, this approach does not take into account the contextual dependencies between each segmented object. To solve this problem, a joint superpixel and Transformer (JST) framework is proposed for HRI classification. In JST, HRI is first segmented into superpixel objects as input, and Transformer is used to model the long-range dependencies. The contextual relationship between each input superpixel object is obtained and the class of analyzed objects is output by designing an encoding and decoding Transformer. Additionally, we explore the effect of semantic range on classification accuracy. JST is also tested by using two HRI datasets with overall classification accuracy, average accuracy and Kappa coefficients of 0.79, 0.70, 0.78 and 0.91, 0.85, 0.89, respectively. The effectiveness of the proposed method is compared qualitatively and quantitatively, and the results achieve competitive and consistently better than the benchmark comparison method | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Deep learning | |
650 | 4 | |a Image classification | |
650 | 4 | |a Remote sensing image | |
650 | 4 | |a Superpixel | |
650 | 4 | |a Transformer | |
700 | 1 | |a Mao, Zhongan |e verfasserin |4 aut | |
700 | 1 | |a Zhang, Tingyu |e verfasserin |4 aut | |
700 | 1 | |a Liu, Tao |e verfasserin |4 aut | |
700 | 1 | |a Wang, Tao |e verfasserin |4 aut | |
700 | 1 | |a Li, Liangzhi |e verfasserin |4 aut | |
700 | 1 | |a Gao, Yu |e verfasserin |4 aut | |
700 | 1 | |a Tian, Runqing |e verfasserin |4 aut | |
700 | 1 | |a Wang, Kun |e verfasserin |4 aut | |
700 | 1 | |a Han, Ling |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Scientific reports |d 2011 |g 14(2024), 1 vom: 01. März, Seite 5054 |w (DE-627)NLM215703936 |x 2045-2322 |7 nnns |
773 | 1 | 8 | |g volume:14 |g year:2024 |g number:1 |g day:01 |g month:03 |g pages:5054 |
856 | 4 | 0 | |u http://dx.doi.org/10.1038/s41598-024-55482-y |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 14 |j 2024 |e 1 |b 01 |c 03 |h 5054 |