Quasi-3D Fluorescence Spectroscopy for Rapid On-Site Algae Species Analysis

This article presents a novel quasi-3D fluorescence spectroscopy technique for rapid and on-site algae species classification. By integrating multiwavelength excitation with a software lock-in amplifier (SLIA) system, the method achieves high signal-to-noise ratio (SNR) and robust fluorescence detection under complex ambient light conditions. Experimental results show that the SNR improved from −9.54 to 14.11 dB at 10 m, and up to 26.43 dB at 5 m. Using four excitation wavelengths (405, 465, 520, and 638 nm), fluorescence spectra were collected from four representative microalgae species, yielding a dataset of 4812 samples. Dimensionality reduction via principal component analysis (PCA) and classification using a lightweight SVM model resulted in 100% test accuracy. Compared to conventional 3-D or discrete spectral methods, the proposed system significantly reduces acquisition time to within 10 s while retaining critical spectral information. The technique shows strong potential for field-deployable applications such as harmful algal bloom detection, aquatic ecosystem monitoring, and real-time environmental sensing. Future work will focus on expanding species coverage and validating performance under dynamic environmental conditions.