Identification of Species and Concentration Measurement of Microalgae Based on Hyperspectral Imaging
Shen Ying,Wu Pan,Huang Feng,Guo Cui-xia
DOI: https://doi.org/10.3964/j.issn.1000-0593(2023)11-3629-08
2023-01-01
Spectroscopy and spectral analysis
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 (R-2) 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 1.05x10(3)similar to 1.05x10(4) and 1.13x10(4)similar to 2.38x10(5) cells center dot 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 1.06x10(5)similar to 4.36x10(6) cells center dot 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.