基于高光谱成像技术的赤潮藻种鉴别和浓度测量方法

doi:10.3964/j.issn.1000-0593(2023)11-3629-08

Abstract
References
Related Citation (15)

Download: PDF (3590 KB)
Export: BibTeX | EndNote (RIS)

Abstract Algal bloom, a water pollution caused by marine algae, may threaten the development of fisheries due to some toxic algal species. Rapid and accurate identification of red tide algal species and their cell concentrations is important for pollution control and management. Traditional detection methods such as microscope and gene sequencing have low timeliness, remote sensing is susceptible to environmental interference resulting in low accuracy, and fluorescence spectroscopy is too expensive for widespread use. Hyperspectral imaging (HSI) technology provides a rapid and non-destructive method for detecting red tide algae species. In this study, a HSI detection system was built to establish a large amount of hyperspectral sample libraries constituted of dinophyta (Amphidinium carterae), bacillariophyta (Skeletonema costatumand Phaeodactylum tricornutum) and raphidophyceae (Heterosigma akashiwo). Two classification methods and three regression methods were used to construct models for algal species identification and cell concentration measurement, respectively, and the effects of seven spectral pretreatment methods (Autoscaling, Normalization, Multiplicative Scatter Correction, Standard Normalized Variate, Savitzky-Golay Smoothing, First Derivative Based on Savitzky-Golay, and Second Derivative Based on Savitzky-Golay) and two band extraction methods (Genetic Algorithms and Successive Projections Algorithm) on the accuracy of modelling were investigated. The results showed that the Second Derivative Based on Savitzky-Golay (SG+2^nd) pretreatment method can improve the accuracy of band extraction and modelling, and that the feature bands selected by the genetic algorithm (GA) are more representative and effective. The feature bands (644.7, 547.8, 562.6, 829.4, 832 nm) extracted SG+2^nd-GA correspond to the absorption spectral bands of specific pigments in the selected algae, combined with Support Vector Machine (SVM) or Back Propagation Neural Network (BPNN) modellingrealized the effective identification of dinophyta, bacillariophyta and raphidophyceae using HSI technology. Compared to Multiple Linear Regression (MLR) and Partial Least Squares (PLS) algorithms, Support Vector Regression (SVR) modelling achieved higher accuracy incell concentration measurements. The coefficients of determination (R²) of the four algal SG+2^nd-GA-SVR cellconcentrations prediction models were all greater than 0.98. Among them, the predicted concentrations of A. carterae e and S. costatumranged from 1.05×10³~1.05×10⁴ and 1.13×10⁴~2.38×10⁵ cells·mL^-1, with the lowest measured concentrations reaching the benchmark concentrations for this algae species in the event of red tide. The predicted concentrations of P. tricornutum ranged from 1.06×10⁵~4.36×10⁶ cells·mL^-1, with the lowest measured concentrations being lower than those of existing spectroscopic techniques. This study provides a new method for rapid, accurate, non-destructive algal blooms detection.

Key words：Algal blooms; Hyperspectral imaging; Species identification; Concentration measurement

Received: 2022-06-10 Accepted: 2022-09-21

ZTFLH:

X55

Corresponding Authors: HUANG Feng E-mail: huangf@fzu.edu.cn

Cite this article:

SHEN Ying,WU Pan,HUANG Feng, et al. Identification of Species and Concentration Measurement of Microalgae Based on Hyperspectral Imaging[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(11): 3629-3636.

URL:

https://www.gpxygpfx.com/EN/10.3964/j.issn.1000-0593(2023)11-3629-08 OR https://www.gpxygpfx.com/EN/Y2023/V43/I11/3629

[1] QI Xiao-li, WU Zhen-zhen, ZHANG Chuan-song, et al(齐晓丽，吴珍珍，张传松，等). Periodical of Ocean University of China[中国海洋大学学报(自然科学版)], 2016, 46(12): 73.
[2] Liu J Y, Zeng L H, Ren Z H. Applied Spectroscopy Reviews, 2021, 56(3): 171.
[3] Santos-Ballardo D U, Rossi S, Hernández V, et al. Aquaculture, 2015, 448: 87.
[4] State Oceanic Administration(国家海洋局). HY/T 069—2005 (2005), Technical Specification for Red Tide Monitoring(HY/T 069—2005，赤潮监测技术规程）, 2005.
[5] Martínez-Guijarro R, Pachés M, Ferrer J, et al. Environmental Technology & Innovation, 2018, 10: 122.
[6] Liu J Y, Zeng L H, Ren Z H, et al. Algal Research, 2020, 45: 101739.
[7] Li X L, Chen K, He Y. Algal Research, 2020, 45: 101680.
[8] Galvao R K H, Araujo M C U, Jose G E, et al. Talanta, 2005, 67(4): 736.
[9] Leardi R, González A L. Chemometrics and Intelligent Laboratory Systems, 1998, 41(2): 195.
[10] Shao Y N, Li Y, Jiang L J, et al. Water Research, 2016, 104: 432.
[11] Baig H S, Saifullah S M, Dar A. Harmful Algae, 2006, 5(2): 133.
[12] Jack Rensel J E, Haigh N, Tynan T J. Harmful Algae, 2010, 10(1): 98.
[13] Li C Q, Zhu B B, Chen H, et al. Journal of Coastal Research, 2009, 25(3): 651.
[14] Morais K C C, Conceição D, Vargas J V C, et al. Renewable Energy, 2021, 163: 138.
[15] Chao Rodríguez Y, Domínguez Gómez J A, Sánchez-Carnero N, et al. Algal Research, 2017, 26: 463.
[16] Clementson L A, Wojtasiewicz B. Data in Brief, 2019, 24: 103875.
[17] ZHAO Dong-zhi(赵冬至). Research Corpus of Redtide Monitoring and Evaluation in Bohai(渤海赤潮灾害监测与评估研究文集). Beijing: Ocean Press(北京：海洋出版社), 2000. 118.
[18] Challagulla V, Walsh K B, Subedi P. Bioenergy Research, 2014, 7(1): 306.
[19] ZHU Hong-yan, SHAO Yong-ni, JIANG Lu-lu, et al(朱红艳，邵咏妮，蒋璐璐，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2016, 36(1): 75.
[20] Rodrigues L H R, Arenzon A, Raya-Rodriguez M T, et al. Journal of Environmental Chemistry and Ecotoxicology, 2011, 3(8): 225.
[21] Catlett D, Siegel D A. Journal of Geophysical Research: Oceans, 2018, 123(1): 246.
[22] Bricaud A, Claustre H, Ras J, et al. Journal of Geophysical Research: Oceans, 2004, 109(C11). doi: 10.1029/2004JC002419.