Aquaculture Areas Extraction Model Using Semantic Segmentation from Remote Sensing Images at the Maowei Sea of Beibu Gulf

The extraction of aquaculture areas from high-resolution remote sensing images is crucial for effective coastal management and resource preservation. This study introduces SwinNet, a semantic segmentation model leveraging multi-scale feature fusion to enhance the extraction of aquaculture areas, par...

Full description

Saved in:
Bibliographic Details
Main Authors: Weirong Qin, Mohd Hasmadi Ismail, Yangyang Luo, Yifeng Yuan, Junlin Deng, Mohammad Firuz Ramli, Ning Wu
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Fishes
Subjects:
Online Access:https://www.mdpi.com/2410-3888/10/5/236
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849327304558247936
author Weirong Qin
Mohd Hasmadi Ismail
Yangyang Luo
Yifeng Yuan
Junlin Deng
Mohammad Firuz Ramli
Ning Wu
author_facet Weirong Qin
Mohd Hasmadi Ismail
Yangyang Luo
Yifeng Yuan
Junlin Deng
Mohammad Firuz Ramli
Ning Wu
author_sort Weirong Qin
collection DOAJ
description The extraction of aquaculture areas from high-resolution remote sensing images is crucial for effective coastal management and resource preservation. This study introduces SwinNet, a semantic segmentation model leveraging multi-scale feature fusion to enhance the extraction of aquaculture areas, particularly in the Maowei Sea of the Beibu Gulf, China. Utilizing the Swin Transformer backbone and a novel Parallel Pooling Attention Module (PPAM), SwinNet minimizes background noise and improves segmentation accuracy. SwinNet achieved a pixel accuracy of 96.53% and an intersection over the union of 93.07% on an aquaculture dataset, demonstrating superior performance in overcoming noise and accurately extracting aquaculture areas. SwinNet offers an effective solution for large-scale, high-precision monitoring of coastal aquaculture, with potential broader applicability in aquatic resource conservation and management.
format Article
id doaj-art-27a1c80ad16b4299a6fa7f1e380895be
institution Kabale University
issn 2410-3888
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Fishes
spelling doaj-art-27a1c80ad16b4299a6fa7f1e380895be2025-08-20T03:47:54ZengMDPI AGFishes2410-38882025-05-0110523610.3390/fishes10050236Aquaculture Areas Extraction Model Using Semantic Segmentation from Remote Sensing Images at the Maowei Sea of Beibu GulfWeirong Qin0Mohd Hasmadi Ismail1Yangyang Luo2Yifeng Yuan3Junlin Deng4Mohammad Firuz Ramli5Ning Wu6Key Laboratory of Beibu Gulf Offshore Engineering Equipment and Technology, Beibu Gulf University, Qinzhou 535011, ChinaFaculty of Forestry and Environment, Universiti Putra Malaysia, Serdang 43400, Selangor, MalaysiaSchool of Computer, Electronics and Information, Guangxi University, Nanning 530004, ChinaKey Laboratory of Beibu Gulf Offshore Engineering Equipment and Technology, Beibu Gulf University, Qinzhou 535011, ChinaKey Laboratory of Beibu Gulf Offshore Engineering Equipment and Technology, Beibu Gulf University, Qinzhou 535011, ChinaFaculty of Forestry and Environment, Universiti Putra Malaysia, Serdang 43400, Selangor, MalaysiaKey Laboratory of Beibu Gulf Offshore Engineering Equipment and Technology, Beibu Gulf University, Qinzhou 535011, ChinaThe extraction of aquaculture areas from high-resolution remote sensing images is crucial for effective coastal management and resource preservation. This study introduces SwinNet, a semantic segmentation model leveraging multi-scale feature fusion to enhance the extraction of aquaculture areas, particularly in the Maowei Sea of the Beibu Gulf, China. Utilizing the Swin Transformer backbone and a novel Parallel Pooling Attention Module (PPAM), SwinNet minimizes background noise and improves segmentation accuracy. SwinNet achieved a pixel accuracy of 96.53% and an intersection over the union of 93.07% on an aquaculture dataset, demonstrating superior performance in overcoming noise and accurately extracting aquaculture areas. SwinNet offers an effective solution for large-scale, high-precision monitoring of coastal aquaculture, with potential broader applicability in aquatic resource conservation and management.https://www.mdpi.com/2410-3888/10/5/236remote sensing imagesaquaculture areassemantic segmentationswin transformermulti-scale feature fusion
spellingShingle Weirong Qin
Mohd Hasmadi Ismail
Yangyang Luo
Yifeng Yuan
Junlin Deng
Mohammad Firuz Ramli
Ning Wu
Aquaculture Areas Extraction Model Using Semantic Segmentation from Remote Sensing Images at the Maowei Sea of Beibu Gulf
Fishes
remote sensing images
aquaculture areas
semantic segmentation
swin transformer
multi-scale feature fusion
title Aquaculture Areas Extraction Model Using Semantic Segmentation from Remote Sensing Images at the Maowei Sea of Beibu Gulf
title_full Aquaculture Areas Extraction Model Using Semantic Segmentation from Remote Sensing Images at the Maowei Sea of Beibu Gulf
title_fullStr Aquaculture Areas Extraction Model Using Semantic Segmentation from Remote Sensing Images at the Maowei Sea of Beibu Gulf
title_full_unstemmed Aquaculture Areas Extraction Model Using Semantic Segmentation from Remote Sensing Images at the Maowei Sea of Beibu Gulf
title_short Aquaculture Areas Extraction Model Using Semantic Segmentation from Remote Sensing Images at the Maowei Sea of Beibu Gulf
title_sort aquaculture areas extraction model using semantic segmentation from remote sensing images at the maowei sea of beibu gulf
topic remote sensing images
aquaculture areas
semantic segmentation
swin transformer
multi-scale feature fusion
url https://www.mdpi.com/2410-3888/10/5/236
work_keys_str_mv AT weirongqin aquacultureareasextractionmodelusingsemanticsegmentationfromremotesensingimagesatthemaoweiseaofbeibugulf
AT mohdhasmadiismail aquacultureareasextractionmodelusingsemanticsegmentationfromremotesensingimagesatthemaoweiseaofbeibugulf
AT yangyangluo aquacultureareasextractionmodelusingsemanticsegmentationfromremotesensingimagesatthemaoweiseaofbeibugulf
AT yifengyuan aquacultureareasextractionmodelusingsemanticsegmentationfromremotesensingimagesatthemaoweiseaofbeibugulf
AT junlindeng aquacultureareasextractionmodelusingsemanticsegmentationfromremotesensingimagesatthemaoweiseaofbeibugulf
AT mohammadfiruzramli aquacultureareasextractionmodelusingsemanticsegmentationfromremotesensingimagesatthemaoweiseaofbeibugulf
AT ningwu aquacultureareasextractionmodelusingsemanticsegmentationfromremotesensingimagesatthemaoweiseaofbeibugulf