A new change detection method for wetlands based on Bi-Temporal Semantic Reasoning UNet++ in Dongting Lake, China
The utility of semantic change detection in myriad change scenarios has garnered considerable attention in contemporary research; however, its applicability in monitoring alterations in wetland ecosystems remains incompletely elucidated. To surmount the constraints associated with binary change dete...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-11-01
|
| Series: | Ecological Indicators |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1470160X23011391 |
| _version_ | 1797655132158033920 |
|---|---|
| author | Yulin Pan Hui Lin Zhuo Zang Jiangping Long Meng Zhang Xiaodong Xu Wenhan Jiang |
| author_facet | Yulin Pan Hui Lin Zhuo Zang Jiangping Long Meng Zhang Xiaodong Xu Wenhan Jiang |
| author_sort | Yulin Pan |
| collection | DOAJ |
| description | The utility of semantic change detection in myriad change scenarios has garnered considerable attention in contemporary research; however, its applicability in monitoring alterations in wetland ecosystems remains incompletely elucidated. To surmount the constraints associated with binary change detection methodologies—chiefly their insufficiency in the extraction of bi-temporal attributes—we introduced the Bi-Temporal Semantic Reasoning UNet++ (Bi-SRUNet++) algorithm. This algorithm leverages the architectural strengths of UNet++ as its foundational network to precisely delineate features pertinent to multi-class change detection. As a preliminary step, the study focused on the Dongting Lake wetland in China and conducted an analysis of feature trends predicated upon the monthly Normalized Difference Water Index (NDWI) and Normalized Difference Vegetation Index (NDVI), as derived from Landsat 8 data spanning 2021–2022. Subsequently, the optimal temporal phases for change detection were ascertained through differential analyses between NDWI and NDVI metrics. Implementing the Bi-SRUNet++ algorithm on a pair of Sentinel-2 images, captured during the optimal phases, yielded augmented change information. Comparative evaluations reveal that the Bi-SRUNet++ algorithm, conceptualized on the framework of Bi-Temporal Semantic Reasoning Network (Bi-SRNet), surpasses the performance indices of its counterpart Semantic Segmentation and Change Detection Late Fusion (SSCD-l). Furthermore, the incorporation of the UNet++ backbone network amplifies the algorithm's capacity for semantic feature extraction, thereby enhancing the efficacy of Bi-SRUNet++ in wetland change detection. The analysis divulges that the total altered area of Dongting Lake during the 2021–2022 period amounts to 1187.97 km2, comprising a water loss of 1186.16 km2, a 715.34 km2 transformation into vegetation, and a conversion of 469.96 km2 into mudflats. The codes and partial dataset in this paper are available at: https://github.com/vivianmiumiu/Bi-SRUNetplusplus-for-SCD. |
| first_indexed | 2024-03-11T17:09:43Z |
| format | Article |
| id | doaj.art-cbe085ef86ef4525afa5b8344892068a |
| institution | Directory Open Access Journal |
| issn | 1470-160X |
| language | English |
| last_indexed | 2024-03-11T17:09:43Z |
| publishDate | 2023-11-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Ecological Indicators |
| spelling | doaj.art-cbe085ef86ef4525afa5b8344892068a2023-10-20T06:38:40ZengElsevierEcological Indicators1470-160X2023-11-01155110997A new change detection method for wetlands based on Bi-Temporal Semantic Reasoning UNet++ in Dongting Lake, ChinaYulin Pan0Hui Lin1Zhuo Zang2Jiangping Long3Meng Zhang4Xiaodong Xu5Wenhan Jiang6Research Center of Forestry Remote Sensing & Information Engineering, Central South University of Forestry & Technology, Changsha 410004, China; Hunan Provincial Key Laboratory of Forestry Remote Sensing Based Big Data & Ecological Security, Changsha 410004, China; Key Laboratory of National Forestry and Grassland Administration on Forest Resources Management and Monitoring in Southern China, Changsha 410004, ChinaResearch Center of Forestry Remote Sensing & Information Engineering, Central South University of Forestry & Technology, Changsha 410004, China; Hunan Provincial Key Laboratory of Forestry Remote Sensing Based Big Data & Ecological Security, Changsha 410004, China; Key Laboratory of National Forestry and Grassland Administration on Forest Resources Management and Monitoring in Southern China, Changsha 410004, ChinaResearch Center of Forestry Remote Sensing & Information Engineering, Central South University of Forestry & Technology, Changsha 410004, China; Hunan Provincial Key Laboratory of Forestry Remote Sensing Based Big Data & Ecological Security, Changsha 410004, China; Key Laboratory of National Forestry and Grassland Administration on Forest Resources Management and Monitoring in Southern China, Changsha 410004, China; Corresponding author at: Research Center of Forestry Remote Sensing & Information Engineering, Central South University of Forestry & Technology, Changsha 410004, China.Research Center of Forestry Remote Sensing & Information Engineering, Central South University of Forestry & Technology, Changsha 410004, China; Hunan Provincial Key Laboratory of Forestry Remote Sensing Based Big Data & Ecological Security, Changsha 410004, China; Key Laboratory of National Forestry and Grassland Administration on Forest Resources Management and Monitoring in Southern China, Changsha 410004, ChinaResearch Center of Forestry Remote Sensing & Information Engineering, Central South University of Forestry & Technology, Changsha 410004, China; Hunan Provincial Key Laboratory of Forestry Remote Sensing Based Big Data & Ecological Security, Changsha 410004, China; Key Laboratory of National Forestry and Grassland Administration on Forest Resources Management and Monitoring in Southern China, Changsha 410004, ChinaResearch Center of Forestry Remote Sensing & Information Engineering, Central South University of Forestry & Technology, Changsha 410004, China; Hunan Provincial Key Laboratory of Forestry Remote Sensing Based Big Data & Ecological Security, Changsha 410004, China; Key Laboratory of National Forestry and Grassland Administration on Forest Resources Management and Monitoring in Southern China, Changsha 410004, China; Guangdong Eco-Engineering Polytechnic, Guangzhou 510520, ChinaResearch Center of Forestry Remote Sensing & Information Engineering, Central South University of Forestry & Technology, Changsha 410004, China; Hunan Provincial Key Laboratory of Forestry Remote Sensing Based Big Data & Ecological Security, Changsha 410004, China; Key Laboratory of National Forestry and Grassland Administration on Forest Resources Management and Monitoring in Southern China, Changsha 410004, ChinaThe utility of semantic change detection in myriad change scenarios has garnered considerable attention in contemporary research; however, its applicability in monitoring alterations in wetland ecosystems remains incompletely elucidated. To surmount the constraints associated with binary change detection methodologies—chiefly their insufficiency in the extraction of bi-temporal attributes—we introduced the Bi-Temporal Semantic Reasoning UNet++ (Bi-SRUNet++) algorithm. This algorithm leverages the architectural strengths of UNet++ as its foundational network to precisely delineate features pertinent to multi-class change detection. As a preliminary step, the study focused on the Dongting Lake wetland in China and conducted an analysis of feature trends predicated upon the monthly Normalized Difference Water Index (NDWI) and Normalized Difference Vegetation Index (NDVI), as derived from Landsat 8 data spanning 2021–2022. Subsequently, the optimal temporal phases for change detection were ascertained through differential analyses between NDWI and NDVI metrics. Implementing the Bi-SRUNet++ algorithm on a pair of Sentinel-2 images, captured during the optimal phases, yielded augmented change information. Comparative evaluations reveal that the Bi-SRUNet++ algorithm, conceptualized on the framework of Bi-Temporal Semantic Reasoning Network (Bi-SRNet), surpasses the performance indices of its counterpart Semantic Segmentation and Change Detection Late Fusion (SSCD-l). Furthermore, the incorporation of the UNet++ backbone network amplifies the algorithm's capacity for semantic feature extraction, thereby enhancing the efficacy of Bi-SRUNet++ in wetland change detection. The analysis divulges that the total altered area of Dongting Lake during the 2021–2022 period amounts to 1187.97 km2, comprising a water loss of 1186.16 km2, a 715.34 km2 transformation into vegetation, and a conversion of 469.96 km2 into mudflats. The codes and partial dataset in this paper are available at: https://github.com/vivianmiumiu/Bi-SRUNetplusplus-for-SCD.http://www.sciencedirect.com/science/article/pii/S1470160X23011391Semantic change detectionSemantic segmentationDongting Lake wetlandBi-SRUNet++ |
| spellingShingle | Yulin Pan Hui Lin Zhuo Zang Jiangping Long Meng Zhang Xiaodong Xu Wenhan Jiang A new change detection method for wetlands based on Bi-Temporal Semantic Reasoning UNet++ in Dongting Lake, China Ecological Indicators Semantic change detection Semantic segmentation Dongting Lake wetland Bi-SRUNet++ |
| title | A new change detection method for wetlands based on Bi-Temporal Semantic Reasoning UNet++ in Dongting Lake, China |
| title_full | A new change detection method for wetlands based on Bi-Temporal Semantic Reasoning UNet++ in Dongting Lake, China |
| title_fullStr | A new change detection method for wetlands based on Bi-Temporal Semantic Reasoning UNet++ in Dongting Lake, China |
| title_full_unstemmed | A new change detection method for wetlands based on Bi-Temporal Semantic Reasoning UNet++ in Dongting Lake, China |
| title_short | A new change detection method for wetlands based on Bi-Temporal Semantic Reasoning UNet++ in Dongting Lake, China |
| title_sort | new change detection method for wetlands based on bi temporal semantic reasoning unet in dongting lake china |
| topic | Semantic change detection Semantic segmentation Dongting Lake wetland Bi-SRUNet++ |
| url | http://www.sciencedirect.com/science/article/pii/S1470160X23011391 |
| work_keys_str_mv | AT yulinpan anewchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT huilin anewchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT zhuozang anewchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT jiangpinglong anewchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT mengzhang anewchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT xiaodongxu anewchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT wenhanjiang anewchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT yulinpan newchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT huilin newchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT zhuozang newchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT jiangpinglong newchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT mengzhang newchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT xiaodongxu newchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina AT wenhanjiang newchangedetectionmethodforwetlandsbasedonbitemporalsemanticreasoningunetindongtinglakechina |