A Synergic Use of Sentinel-1 and Sentinel-2 Imagery for Complex Wetland Classification Using Generative Adversarial Network (GAN) Scheme

被引:11
|
作者
Jamali, Ali [1 ]
Mahdianpari, Masoud [2 ,3 ]
Mohammadimanesh, Fariba [3 ]
Brisco, Brian [4 ]
Salehi, Bahram [5 ]
机构
[1] Univ Karabuk, Fac Engn, Civil Engn Dept, TR-78050 Karabuk, Turkey
[2] Mem Univ Newfoundland, Dept Elect & Comp Engn, St John, NF A1B 3X5, Canada
[3] C CORE, 1 Morrissey Rd, St John, NF A1B 3X5, Canada
[4] Canada Ctr Mapping & Earth Observat, Ottawa, ON K1S 5K2, Canada
[5] SUNY Syracuse, Coll Environm Sci & Forestry SUNY ESF, Dept Environm Resources Engn, Syracuse, NY 13210 USA
关键词
wetland classification; machine learning; CNN; Deep Convolutional Neural Network; Generative Adversarial Network; OBJECT DETECTION; SATELLITE;
D O I
10.3390/w13243601
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Due to anthropogenic activities and climate change, many natural ecosystems, especially wetlands, are lost or changing at a rapid pace. For the last decade, there has been increasing attention towards developing new tools and methods for the mapping and classification of wetlands using remote sensing. At the same time, advances in artificial intelligence and machine learning, particularly deep learning models, have provided opportunities to advance wetland classification methods. However, the developed deep and very deep algorithms require a higher number of training samples, which is costly, logistically demanding, and time-consuming. As such, in this study, we propose a Deep Convolutional Neural Network (DCNN) that uses a modified architecture of the well-known DCNN of the AlexNet and a Generative Adversarial Network (GAN) for the generation and classification of Sentinel-1 and Sentinel-2 data. Applying to an area of approximately 370 sq. km in the Avalon Peninsula, Newfoundland, the proposed model with an average accuracy of 92.30% resulted in F-1 scores of 0.82, 0.85, 0.87, 0.89, and 0.95 for the recognition of swamp, fen, marsh, bog, and shallow water, respectively. Moreover, the proposed DCNN model improved the F-1 score of bog, marsh, fen, and swamp wetland classes by 4%, 8%, 11%, and 26%, respectively, compared to the original CNN network of AlexNet. These results reveal that the proposed model is highly capable of the generation and classification of Sentinel-1 and Sentinel-2 wetland samples and can be used for large-extent classification problems.
引用
收藏
页数:16
相关论文
共 50 条
  • [1] Swin Transformer for Complex Coastal Wetland Classification Using the Integration of Sentinel-1 and Sentinel-2 Imagery
    Jamali, Ali
    Mahdianpari, Masoud
    [J]. WATER, 2022, 14 (02)
  • [2] META-LEARNING FOR WETLAND CLASSIFICATION USING A COMBINATION OF SENTINEL-1 AND SENTINEL-2 IMAGERY
    Jafarzadeh, Hamid
    Mahdianpari, Masoud
    Gill, Eric
    [J]. XXIV ISPRS CONGRESS: IMAGING TODAY, FORESEEING TOMORROW, COMMISSION III, 2022, 5-3 : 47 - 52
  • [3] WETLAND CLASSIFICATION WITH SWIN TRANSFORMER USING SENTINEL-1 AND SENTINEL-2 DATA
    Jamali, Ali
    Mohammadimanesh, Fariba
    Mahdianpari, Masoud
    [J]. 2022 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2022), 2022, : 6213 - 6216
  • [4] Classification scheme for mapping wetland herbaceous plant communities using time series Sentinel-1 and Sentinel-2 data
    Zhang, Li
    Luo, Wenting
    Zhang, Haohuan
    Yin, Xiuwan
    Li, Bin
    [J]. National Remote Sensing Bulletin, 2023, 27 (06) : 6 - 19
  • [5] Improving Urban Land Cover Classification with Combined Use of Sentinel-2 and Sentinel-1 Imagery
    Hu, Bin
    Xu, Yongyang
    Huang, Xiao
    Cheng, Qimin
    Ding, Qing
    Bai, Linze
    Li, Yan
    [J]. ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION, 2021, 10 (08)
  • [6] DEEP RECURRENT NEURAL NETWORK FOR CROP CLASSIFICATION TASK BASED ON SENTINEL-1 AND SENTINEL-2 IMAGERY
    Kussul, Nataliia
    Lavreniuk, Mykola
    Shumilo, Leonid
    [J]. IGARSS 2020 - 2020 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, 2020, : 6914 - 6917
  • [7] Reconstructing Snow-Free Sentinel-2 Satellite Imagery: A Generative Adversarial Network (GAN) Approach
    Oluwadare, Temitope Seun
    Chen, Dongmei
    Oluwafemi, Olawale
    Babadi, Masoud
    Hossain, Mohammad
    Ibukun, Oluwaseun
    [J]. REMOTE SENSING, 2024, 16 (13)
  • [8] Super-Resolution of Sentinel-2 Imagery Using Generative Adversarial Networks
    Salgueiro Romero, Luis
    Marcello, Javier
    Vilaplana, Veronica
    [J]. REMOTE SENSING, 2020, 12 (15)
  • [9] Synergy of Sentinel-1 and Sentinel-2 Imagery for Crop Classification Based on DC-CNN
    Zhang, Kaixin
    Yuan, Da
    Yang, Huijin
    Zhao, Jianhui
    Li, Ning
    [J]. REMOTE SENSING, 2023, 15 (11)
  • [10] Mountain crop monitoring with multitemporal Sentinel-1 and Sentinel-2 imagery
    Notarnicola, C.
    Asam, S.
    Jacob, A.
    Marin, C.
    Rossi, M.
    Stendardi, L.
    [J]. 2017 9TH INTERNATIONAL WORKSHOP ON THE ANALYSIS OF MULTITEMPORAL REMOTE SENSING IMAGES (MULTITEMP), 2017,