Semantic Segmentation of Marine Remote Sensing Based on a Cross Direction Attention Mechanism
With the development of remote sensing technology, the semantic segmentation and recognition of various things in the ocean have become more and more frequent. Due to the wide variety of marine things and the large differences in morphology, it has brought greater difficulties to the recognition of...
Published in: | IEEE Access |
---|---|
Main Authors: | , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
IEEE
2020
|
Subjects: | |
Online Access: | https://doi.org/10.1109/ACCESS.2020.3013898 https://doaj.org/article/ee376cedfdd74e539fba258f1308e301 |
id |
ftdoajarticles:oai:doaj.org/article:ee376cedfdd74e539fba258f1308e301 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:ee376cedfdd74e539fba258f1308e301 2023-05-15T18:18:42+02:00 Semantic Segmentation of Marine Remote Sensing Based on a Cross Direction Attention Mechanism Hao Gao Lin Cao Dingfeng Yu Xuejun Xiong Maoyong Cao 2020-01-01T00:00:00Z https://doi.org/10.1109/ACCESS.2020.3013898 https://doaj.org/article/ee376cedfdd74e539fba258f1308e301 EN eng IEEE https://ieeexplore.ieee.org/document/9154711/ https://doaj.org/toc/2169-3536 2169-3536 doi:10.1109/ACCESS.2020.3013898 https://doaj.org/article/ee376cedfdd74e539fba258f1308e301 IEEE Access, Vol 8, Pp 142483-142494 (2020) Cross direction attention mechanism marine remote sensing multi-access convolutional deep learning convolution and dilated convolution Electrical engineering. Electronics. Nuclear engineering TK1-9971 article 2020 ftdoajarticles https://doi.org/10.1109/ACCESS.2020.3013898 2022-12-31T15:43:16Z With the development of remote sensing technology, the semantic segmentation and recognition of various things in the ocean have become more and more frequent. Due to the wide variety of marine things and the large differences in morphology, it has brought greater difficulties to the recognition of marine remote sensing images. In order to obtain better segmentation results of ocean remote sensing images, this paper proposes an cross attention mechanism(Horizontal and Vertical) of exponential operation combined with multi-scale convolution algorithm. Among them, the cross attention mechanism and expanded distribution weight coefficient mentioned in this paper are first proposed. First, Input the marine remote sensing image features into an cross attention mechanism algorithm of exponential operation to obtain feature weight coefficients and joint weight coefficients in multiple directions; Then, the features with weight coefficients are input into the multi-access convolutional layer and the multi-scale dilated convolutional layer respectively for deep feature mining; Then the above steps are repeated twice, and finally the semantic segmentation of marine remote sensing images is achieved by fusing multiple deep-level features afterwards. Experiments were conducted on three public marine remote sensing data sets, and the results proved the effectiveness of our proposed cross attention mechanism of extended operation algorithm. The F values of the MAMC model on Beach, Island and Sea ice data sets have reached 99.4%, 91.25%, 87.08% respectively. Compared with other models, the effect is significantly improved, and proved the powerful performance of the algorithm in the semantic segmentation of marine remote sensing images. Article in Journal/Newspaper Sea ice Directory of Open Access Journals: DOAJ Articles Beach Island ENVELOPE(-79.050,-79.050,57.500,57.500) IEEE Access 8 142483 142494 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
Cross direction attention mechanism marine remote sensing multi-access convolutional deep learning convolution and dilated convolution Electrical engineering. Electronics. Nuclear engineering TK1-9971 |
spellingShingle |
Cross direction attention mechanism marine remote sensing multi-access convolutional deep learning convolution and dilated convolution Electrical engineering. Electronics. Nuclear engineering TK1-9971 Hao Gao Lin Cao Dingfeng Yu Xuejun Xiong Maoyong Cao Semantic Segmentation of Marine Remote Sensing Based on a Cross Direction Attention Mechanism |
topic_facet |
Cross direction attention mechanism marine remote sensing multi-access convolutional deep learning convolution and dilated convolution Electrical engineering. Electronics. Nuclear engineering TK1-9971 |
description |
With the development of remote sensing technology, the semantic segmentation and recognition of various things in the ocean have become more and more frequent. Due to the wide variety of marine things and the large differences in morphology, it has brought greater difficulties to the recognition of marine remote sensing images. In order to obtain better segmentation results of ocean remote sensing images, this paper proposes an cross attention mechanism(Horizontal and Vertical) of exponential operation combined with multi-scale convolution algorithm. Among them, the cross attention mechanism and expanded distribution weight coefficient mentioned in this paper are first proposed. First, Input the marine remote sensing image features into an cross attention mechanism algorithm of exponential operation to obtain feature weight coefficients and joint weight coefficients in multiple directions; Then, the features with weight coefficients are input into the multi-access convolutional layer and the multi-scale dilated convolutional layer respectively for deep feature mining; Then the above steps are repeated twice, and finally the semantic segmentation of marine remote sensing images is achieved by fusing multiple deep-level features afterwards. Experiments were conducted on three public marine remote sensing data sets, and the results proved the effectiveness of our proposed cross attention mechanism of extended operation algorithm. The F values of the MAMC model on Beach, Island and Sea ice data sets have reached 99.4%, 91.25%, 87.08% respectively. Compared with other models, the effect is significantly improved, and proved the powerful performance of the algorithm in the semantic segmentation of marine remote sensing images. |
format |
Article in Journal/Newspaper |
author |
Hao Gao Lin Cao Dingfeng Yu Xuejun Xiong Maoyong Cao |
author_facet |
Hao Gao Lin Cao Dingfeng Yu Xuejun Xiong Maoyong Cao |
author_sort |
Hao Gao |
title |
Semantic Segmentation of Marine Remote Sensing Based on a Cross Direction Attention Mechanism |
title_short |
Semantic Segmentation of Marine Remote Sensing Based on a Cross Direction Attention Mechanism |
title_full |
Semantic Segmentation of Marine Remote Sensing Based on a Cross Direction Attention Mechanism |
title_fullStr |
Semantic Segmentation of Marine Remote Sensing Based on a Cross Direction Attention Mechanism |
title_full_unstemmed |
Semantic Segmentation of Marine Remote Sensing Based on a Cross Direction Attention Mechanism |
title_sort |
semantic segmentation of marine remote sensing based on a cross direction attention mechanism |
publisher |
IEEE |
publishDate |
2020 |
url |
https://doi.org/10.1109/ACCESS.2020.3013898 https://doaj.org/article/ee376cedfdd74e539fba258f1308e301 |
long_lat |
ENVELOPE(-79.050,-79.050,57.500,57.500) |
geographic |
Beach Island |
geographic_facet |
Beach Island |
genre |
Sea ice |
genre_facet |
Sea ice |
op_source |
IEEE Access, Vol 8, Pp 142483-142494 (2020) |
op_relation |
https://ieeexplore.ieee.org/document/9154711/ https://doaj.org/toc/2169-3536 2169-3536 doi:10.1109/ACCESS.2020.3013898 https://doaj.org/article/ee376cedfdd74e539fba258f1308e301 |
op_doi |
https://doi.org/10.1109/ACCESS.2020.3013898 |
container_title |
IEEE Access |
container_volume |
8 |
container_start_page |
142483 |
op_container_end_page |
142494 |
_version_ |
1766195368855339008 |