Chlorophyll a inversion algorithm and algae classification technique based on hyperspectral data

Accurate monitoring and classification of algal species, as well as the estimation of chlorophyll-a concentration, are crucial for assessing water quality and managing aquatic ecosystems. Algal blooms, particularly those involving toxic species, pose significant threats to water resources, aquatic life, and human health. Traditional remote sensing methods lack the precision and specificity needed to address these challenges effectively. This study aims to develop a more accurate, flexible, and comprehensive approach to chlorophyll-a concentration inversion using hyperspectral data By leveraging the rich spectral information provided by hyperspectral data and combining Long Short-Term Memory (LSTM) networks with one-dimensional Convolutional Neural Networks (1D-CNN), we significantly improve the accuracy of chlorophyll-a concentration inversion. Preliminary measurements in our study area showed chlorophyll-a concentrations exceeding $30 \mu \mathrm{g} / \mathrm{L}$, indicating a eutrophic state. Experimental results demonstrate that this combined approach enhances the ${\mathrm {R}}^{2}$ Coefficient of Determination) between predicted and actual chlorophyll-a concentrations from 0.71 to 0.94, outperforming traditional models. This advancement indicates a significant improvement in aquatic remote sensing technology, providing a more precise and reliable method for water quality monitoring and understanding aquatic ecosystem dynamics.