基于多尺度注意力融合的叶绿素a水质参数反演研究

doi:10.3964/j.issn.1000-0593(2025)04-1190-11

Abstract
References
Related Citation (15)

Download: PDF (17630 KB)
Export: BibTeX | EndNote (RIS)

Abstract Water resources are one of the core elements of the ecological environment, and there is currently a large number of water bodies being polluted by industrialization or nutrient enrichment, making real-time monitoring of water quality parameters crucial for maintaining the health of water bodies. Traditional water quality monitoring methods often use on-site sampling measurement or linear regression prediction methods. Still, due to the significant non-linear characteristics of water quality parameters, traditional water quality monitoring methods are time-consuming and inaccurate in their predictions. In recent years, deep learning methods have shown good performance in dealing with complex non-linear problems and have been applied by many scholars to the inverse estimation of water quality parameters. However, water quality inversion models based on deep learning still suffer from inaccurate analysis of remote sensing spectral images and poor model generalization capabilities. Therefore, this paper proposes a water quality inversion network model based on a multiscale attention fusion mechanism, which can accurately predict water quality parameters such as chlorophyll-a, providing a basis for assessing the health of water bodies. The network integrates advanced attention mechanisms and feature fusion strategies, combining the advantages of local feature learning from CNN and global feature extraction capabilities from Transformer to construct a Dense ASPP module for obtaining multiscale features of remote sensing images. It uses a channel attention decoder and pooling fusion modules to extract detailed features. Then, it estimates the concentration of chlorophyll-a by fusing different scales and levels of feature information, achieving higher inversion accuracy and generalization performance. To validate the performance of the proposed inversion model, experiments were implemented in Python 3.7 and the PyTorch framework, using ocean remote sensing spectral images and chlorophyll-a concentration data from January 2021 to December 2022 for network training. The experiment compares the proposed method with seven other water quality inversion methods, achieving the best performance in all objective indicators. It improves the R² index by 0.09 compared to the best method in the comparison, and reduces the RMSE, MAE and MAD indices by 11.99, 0.089, and 0.029, respectively, and improves the E_var index by 0.098, and the NSE index by 0.041. Meanwhile, in subjective evaluation, the proposed method obtains more precise chlorophyll-a concentrations, smaller errors, and higher generalization ability in different waters.

Key words：Water quality inversion; Chlorophyll-a; Remote sensing spectral images; Multiscale attention fusion

Received: 2024-05-07 Accepted: 2024-08-14

ZTFLH:	X87
	X832

Corresponding Authors: YU Tao E-mail: yutao@opt.ac.cn

Cite this article:

SUN Bang-yong,GONG Kai-jie,YU Tao, et al. Research on Chlorophyll-a Water Quality Parameter Inversion Based on Multi-Scale Attention Fusion Network Model[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(04): 1190-1200.

URL:

https://www.gpxygpfx.com/EN/10.3964/j.issn.1000-0593(2025)04-1190-11 OR https://www.gpxygpfx.com/EN/Y2025/V45/I04/1190

[1] ZHOU Li-hua, LIU Yang(周立华，刘洋). Acta EcologicaSinica(生态学报), 2021, (8): 3306.
[2] LI Zheng-hao, CHEN Zhi-zhao, WANG Li-yan, et al(李正浩, 陈志钊, 王力彦, 等). Acta Optica Analysis(光学学报), 2021, 41(2): 7.
[3] WANG Si-meng, QIN Bo-qiang(王思梦，秦伯强). Environmental Science(环境科学), 2022, 44(3): 1228.
[4] Cao Jiaju, Wen Xingping, Luo Dayou, et al. Journal of Spectroscopy，2022：394.
[5] ZHANG Yi-ting, LI Tian-hong(张贻婷, 李天宏). Environmental Science & Technology(环境科学与技术), 2021，(8): 044.
[6] Pyo J C, Duan H, Baek S, et al. Remote Sensing of Environment, 2019, 233: 111350.
[7] LUO Jie-chunyi, QIN Long-jun, MAO Peng, et al(罗婕纯一, 秦龙君, 毛鹏, 等). Remote Sensing Technology and Application(遥感技术与应用), 2021, 36(3): 473.
[8] TIAN Jian-lin, DENG Ru-ru, QIN Yan, et al(田建林, 邓孺孺, 秦雁, 等). Economic Geography(经济地理), 2018, 38(8): 172.
[9] Hoge F E, Lyon P E. Journal of Geophysical Research: Oceans, 1996, 101(C7): 16631.
[10] O'Reilly J E, Maritorena S, O'Brien M C, et al. NASA Technical Memorandum—SeaWI FS Postlaunch Technical Report Series, 2000, 55(27): 1.
[11] Mishra S, Mishra D R. Remote Sensing of Environment, 2012, 117: 394.
[12] FENG Jia-cheng, LI Yong, LI Na, et al(冯家成, 李勇, 李娜, 等). Environmental Engineering(环境工程), 2022, 40(1): 161.
[13] LI Ao, FENG Yang, WANG Yun-tao, et al(李傲, 冯洋, 王云涛, 等). Journal of Tropical Oceanography(热带海洋学报), 2022, 41(2): 77.
[14] SHI Sui-xiang, WANG Lei, YU Xuan, et al(石绥祥, 王蕾, 余璇, 等). Journal of Oceanography(海洋学报), 2020, 42(2): 134.
[15] ZHOU Yi, LIU De-er(周羿, 刘德儿). Remote Sensing Information(遥感信息)，2023，38(3): 85.
[16] DONG Qian-yu, YANG Qiu-xiang, ZHAO Yin(董倩瑜，杨秋翔，赵印). Laser & Optoelectronics Progrees(激光与光电子学进展)，2025, 1: 1.
[17] MA Zhen, ZHOU Mi(马振, 周密). Journal of China Hydrology(水文), 2018, 38(1): 77.
[18] PAN Ying-yang, GUO Qiao-zhen, FU Ying(潘应阳, 国巧真, 付盈). Environmental Monitoring in China(中国环境监测), 2018, 34(3): 148.
[19] TAO Ran, PENG Jin-chan, ZHANG Hao, et al(陶然, 彭金婵, 张豪, 等). Geomatics World(地理信息世界), 2019, 26(4): 44.
[20] PAN Ying-yang, GUO Qiao-zhen, SUN Jin-hua(潘应阳, 国巧真, 孙金华). Science of Surveying and Mapping(测绘科学), 2017, 42(1): 43.
[21] XIAO Qian, YANG Kun, HONG Liang(肖茜, 杨昆, 洪亮). Journal of Yunnan Normal University(Natural Sciences Edition)[云南师范大学学报(自然科学版)], 2018, 38(2): 73.
[22] Zhang H, Ren X, Chen S, et al. Environmental Pollution, 2024, 0269: 7491.
[23] PANG Ji-yu, ZHANG An-bing, WANG He-feng, et al(庞吉玉, 张安兵, 王贺封, 等). China Rural Water and Hydropower(中国农村水利水电), 2023, (3): 111.
[24] Mohammad G Zamani, Mohammad Reza Nikoo, Ghazi Al-Rawas, et al. Journal of Environmental Management, 2024, 358: 120756.
[25] Wang Xuefei, Wang Tingkai, Li Jiale. Engineering Applications of Artificial Intelligence, 2023: 0952: 1976.