浒苔绿潮遥感监测技术研究进展

doi:10.3976/j.issn.1002-4026.2025150

摘要/Abstract

摘要： 浒苔绿潮是全球规模最大的绿潮灾害，自2007年起已连续19年暴发，该灾害破坏海洋生态平衡、阻碍渔业航运、影响滨海旅游业发展，造成了巨大的经济损失。卫星遥感技术以其大面积、长时序、可回溯等优点，成为浒苔绿潮监测的重要技术手段。浒苔绿潮遥感监测技术的发展主要由算法模型推动，经历了从简单线性模型（光谱指数）到复杂非线性模型（机器学习），再到深层特征表达模型（深度学习）的跨越。光谱指数法依赖预设阈值，易受云层、太阳耀斑等环境因素干扰，机器学习方法需人工筛选特征，在处理复杂遥感图像时精度与效率欠佳，相比之下，深度学习模型展现出显著优势。但目前深度学习在浒苔监测领域的研究尚处于起步阶段，面临极稀疏分布和边缘过度区识别精度不高等问题。未来需通过多源遥感数据融合、提升模型泛化能力、构建标准化样本库等手段，突破现有技术瓶颈，最终实现浒苔绿潮的高精度、智能化监测与预警，为海洋生态保护与区域可持续发展提供技术支撑。

关键词: 浒苔绿潮, 光谱指数, 卫星遥感, 深度学习, 数据融合

Abstract: The Ulva prolifera green tides in the Yellow Sea are the largest-scale green tide disasters in the world. Since 2007, they have occurred continuously for 19 years. These events disrupt the marine ecological balance, hinder fisheries and maritime transportation, and affect the development of coastal tourism, thereby causing substantial economic losses. Satellite remote sensing technology, with its advantages of large spatial coverage, long time series, and retrospective capability, has become an important tool for monitoring Ulva prolifera green tides. The advancement of remote sensing monitoring technology for Ulva prolifera green tides has been primarily driven by algorithmic models, evolving from simple linear models (spectral indices) to complex nonlinear models (machine learning), and further to deep feature representation models (deep learning). Spectral index methods rely on predefined thresholds and are susceptible to interference from environmental factors such as clouds and sun glint. Machine learning methods require manual feature selection, resulting in limited accuracy and efficiency when processing complex remote sensing images. In contrast, deep learning models demonstrate remarkable advantages. However, current research on deep learning in the field of Ulva prolifera monitoring is still in its early stages, facing challenges such as the extremely sparse distribution and low recognition accuracy in transitional boundary regions. Future efforts should focus on multisource remote sensing data fusion, enhancing model generalization capability, and constructing standardized sample databases to overcome current technical bottlenecks. Ultimately, this will enable high-precision, real-time, and intelligent monitoring and early warning of Ulva prolifera green tides, providing technical support for marine ecological protection and sustainable regional development.

Key words: Ulva prolifera green tides, spectral index, satellite remote sensing, deep learning, data fusion

中图分类号:

TP79

李慧平, 禹定峰, 纪泽禄, 杨雷. 浒苔绿潮遥感监测技术研究进展[J]. 山东科学, 0, (): 1-.

LI Huiping, YU Dingfeng, JI Zelu, YANG Lei. Research progress on remote sensing monitoring technology for Ulva Prolifera green tides[J]. Shandong Science, 0, (): 1-.

参考文献

[1] 郭继强, 潘洁, 刘文雅, 等. 基于Suomi NPP VIIRS数据的黄海浒苔遥感监测研究[J]. 安徽农业科学, 2018, 46(15): 63-66. DOI:10.13989/j.cnki.0517-6611.2018.15.020.

[2] 秦泉, 胡先锋, 李峰, 等. 面向国产高分辨率卫星的浒苔绿潮遥感监测算法对比[J]. 海洋气象学报, 2023, 43(2): 64-75. DOI:10.19513/j.cnki.issn2096-3599.2023.02.006.

[3] Gao Le, Li Xiaofeng, Kong Fanzhou, et al. AlgaeNet: A deep-learning framework to detect floating green algae from optical and SAR imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, 15: 2782-2796. DOI:10.1109/JSTARS.2022.3162387.

[4] An Deyu, Yu Dingfeng, Zheng Xiangyang, et al. Monitoring the dissipation of the floating green macroalgae blooms in the Yellow Sea (2007–2020) on the basis of satellite remote sensing[J]. Remote Sensing, 2021, 13(19): 3811. DOI:10.3390/rs13193811.

[5] 王怡人, 王胜强, 喻樾, 等. 一种提取南黄海浒苔的自适应阈值遥感算法[J]. 遥感信息, 2021, 36(2): 120-129. DOI:10.3969/j.issn.1000-3177.2021.02.018.

[6] 万剑华, 万献慈, 刘善伟, 等. 多指数决策融合的MODIS浒苔提取方法[J]. 海洋环境科学, 2022, 41(2): 316-324. DOI:10.13634/j.cnki.mes.2022.02.017.

[7] 董京铭, 石轩硕, 张银意, 等. 基于国产静止高分卫星GF4-MSS数据的浒苔灾害监测研究[J]. 湖北农业科学, 2022, 61(23): 190-196. DOI:10.14088/j.cnki.issn0439-8114.2022.23.038.

[8] 麻德明, 张秀林, 田梓文, 等. 无人机遥感影像浒苔信息提取方法[J]. 测绘通报, 2023(8): 24-28. DOI:10.13474/j.cnki.11-2246.2023.0226.

[9] Wu Shunyao, Shao Fengjing, Wang Ying, et al.Enteromorphaproliferadetection with MODIS image using semi-supervised clustering[J]. Journal of Computers, 2014, 9(5): 1259-1265.DOI:10.4304/jcp.9.5.1259-1265.

[10] 唐鹏飞, 杜培军, 郭山川, 等. 基于Sentinel-1 SAR影像的南黄海浒苔提取与动态监测[J]. 遥感学报, 2024, 28(8): 2030-2044. DOI:10.11834/jrs.20242200.

[11] Zhang Baowei, Guo Jianzhong, Li Ziwei, et al. Identifying the spatio-temporal variations ofUlvaproliferadisasters in all life cycle[J]. Journal of Water and Climate Change, 2022, 13(2): 629-644.

[12] 张海龙, 孙德勇, 李俊生, 等. 基于GF1-WFV和HJ-CCD数据的我国近海绿潮遥感监测算法研究[J]. 光学学报, 2016, 36(6): 601004. DOI:10.3788/AOS201636.0601004.

[13] 陈莹, 孙德勇, 张海龙, 等. 结合GOCI数据的黄海绿潮遥感监测及漂移轨迹研究[J]. 光学学报, 2020, 40(3): 0301001. DOI:10.3788/AOS202040.0301001.

[14] Zhang Hailong, Qiu Zhongfeng, Devred E, et al. A simple and effective method for monitoring floating green macroalgae blooms: A case study in the Yellow Sea[J]. Optics Express, 2019, 27(4): 4528-4548. DOI:10.1364/OE.27.004528.

[15] Wang Changying, Chu Jialan, Tan Meng, et al. An automatic detection of green tide using multi-windows with their adaptive threshold from Landsat TM/ETM plus image[J]. Acta Oceanologica Sinica, 2017, 36(11): 106-114. DOI:10.1007/s13131-017-1141-9.

[16] Xu Shuwen, Yu Tan, Xu Jinmeng, et al. Monitoring and forecasting green tide in the Yellow Sea using satellite imagery[J]. Remote Sensing, 2023, 15(8): 2196. DOI:10.3390/rs15082196.

[17] Li Lin, Zheng Xiangyang, Wei Zhenning, et al. A spectral-mixing model for estimating sub-pixel coverage of sea-surface floating macroalgae[J]. Atmosphere-Ocean, 2018, 56(4): 296-302. DOI:10.1080/07055900.2018.1509834.

[18] Miura Y, Imamoto H, Asada Y, et al. Prediction of algal bloom using a combination of sparse modeling and a machine learning algorithm: Automatic relevance determination and support vector machine[J]. Ecological Informatics, 2023, 78: 102337. DOI:10.1016/j.ecoinf.2023.102337.

[19] Xia Jingjing, Zeng Jin. Early warning of algal blooms based on the optimization support vector machine regression in a typical tributary bay of the Three Gorges Reservoir, China[J]. Environmental Geochemistry and Health, 2022, 44(12): 4719-4733. DOI:10.1007/s10653-022-01203-1.

[20] Zheng Longxiao, Wu Mengquan, Xue Mingyue, et al. Power of SAR imagery and machine learning in monitoringUlvaprolifera: A case study of sentinel-1 and random forest[J]. Chinese Geographical Science, 2024, 34(6): 1134-1143. DOI:10.1007/s11769-024-1465-2.

[21] Hou Wenlong, Chen Jinyue, He Maoxia, et al. Evolutionary trends and analysis of the driving factors ofUlvaproliferagreen tides: A study based on the random forest algorithm and multisource remote sensing images[J]. Marine Environmental Research, 2024, 198: 106495. DOI:10.1016/j.marenvres.2024.106495.

[22] Xiao Yanfang, Liu Rongjie, Kim K, et al. A random forest-based algorithm to distinguishUlvaproliferaandSargassumfrom multispectral satellite images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 4201515. DOI:10.1109/TGRS.2021.3071154.

[23] Geng Xiaomeng, Li Pingxiang, Yang Jie, et al.Ulvaproliferadetection with dual-polarization GF-3 SAR data[J]. IOP Conference Series: Earth and Environmental Science, 2020, 502(1): 012026. DOI:10.1088/1755-1315/502/1/012026.

[24] Qiu Zhongfeng, Li Zhaoxin, Bilal M, et al. Automatic method to monitor floating macroalgae blooms based on multilayer perceptron: Case study of Yellow Sea using GOCI images[J]. Optics Express, 2018, 26(21): 26810-26829. DOI:10.1364/OE.26.026810.

[25] He Kaiming, Zhang Xiangyu, Ren Shaoqing, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA. IEEE, 2016: 770-778. DOI:10.1109/CVPR.2016.90.

[26] Ronneberger O, Fischer P, Brox T. U-Net: Convolutional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015. Cham: Springer, 2015: 234-241. DOI:10.1007/978-3-319-24574-4_28.

[27] Dong Zhipeng, Liu Yanxiong, Wang Yanli, et al.Enteromorphaproliferadetection in high-resolution remote sensing imagery based on boundary-assisted dual-path convolutional neural networks[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 4208715. DOI:10.1109/TGRS.2023.3326500.

[28] Zou Yibo, Wang Xiaoliang, Wang Lei, et al. A high-quality instance-segmentation network for floating-algae detection using RGB images[J]. Remote Sensing, 2022, 14(24): 6247. DOI:10.3390/rs14246247.

[29] Cui Binge, Zhang Haoqing, Jing Wei, et al. SRSe-net: Super-resolution-based semantic segmentation network for green tide extraction[J]. Remote Sensing, 2022, 14(3): 710. DOI:10.3390/rs14030710.

[30] Yin Lirong, Wang Lei, Lu Siyu, et al. Convolution-transformer for image feature extraction[J]. Computer Modeling in Engineering & Sciences, 2024, 141(1): 87-106. DOI:10.32604/cmes.2024.051083.

[31] Dosovitskiy A, Beyer L, Kolesnikov A, et al. An image is worth 16x16 words: Transformers for image recognition at scale[PP/OL]. V2. arXiv (2021-06-03). https://doi.org/10.48550/arXiv.2010.11929.

[32] Patil G A, B S K, Kulkarni P. Detection and classification of harmful algae using machine learning[C]//2025 4th OPJU International Technology Conference (OTCON) on Smart Computing for Innovation and Advancement in Industry 5.0. Raigarh, India. IEEE, 2025: 1-10. DOI:10.1109/OTCON65728.2025.11070713.

[33] Zhu Shengyuan, Wu Yinglei, Ma Xiaoshuang. Deep learning-based algal bloom identification method from remote sensing images—Take China’s Chaohu Lake as an example[J]. Sustainability, 2023, 15(5): 4545. DOI:10.3390/su15054545.

[34] Li Xiaofeng, Liu Bin, Zheng Gang, et al. Deep-learning-based information mining from ocean remote-sensing imagery[J]. National Science Review, 2020, 7(10): 1584-1605. DOI:10.1093/nsr/nwaa047.

[35] Wan Xianci, Wan Jianhua, Xu Mingming, et al.Enteromorphacoverage information extraction by 1D-CNN and Bi-LSTM networks considering sample balance from GOCI images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 9306-9317. DOI:10.1109/JSTARS.2021.3110854.

[36] Wang Zhongyuan, Fang Zhixiang, Wu Yichen, et al. Multi-source evidence data fusion approach to detect daily distribution and coverage ofUlvaProliferain the Yellow Sea, China[J]. IEEE Access, 2019, 7: 115214-115228. DOI:10.1109/ACCESS.2019.2936247.

[37] Jin Xifang, Li Yun, Wan Jianhua, et al. MODIS green-tide detection with a squeeze and excitation oriented generative adversarial network[J]. IEEE Access, 2022, 10: 60294-60305. DOI:10.1109/ACCESS.2022.3180331.

[38] Sheng Hui, Jia Manman, Wei Shiqing, et al. UPE-net: A spatial information enhanced network forUlvaproliferaextraction from GF-1 WFV images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2025, 18: 1946-1962. DOI:10.1109/JSTARS.2024.3492533.