Three-dimensional (3D) fluorescence spectroscopy is a widely used method for collecting fluorescence signals from phytoplankton and is extensively applied in ecological monitoring and environmental research. However, traditional 3D fluorescence spectroscopy faces challenges in efficiency and real-time performance for on-site rapid detection. In this paper, we propose a parallel acquisition method for quasi-3D fluorescence spectra using software lock-in amplification technique to address these issues. By employing heterodyne modulation of discrete multi-wavelength excitation sources and integrating high-resolution, broad-spectrum signal acquisition with software lock-in amplification, the method enables the simultaneous detection and separation of multi-wavelength-induced fluorescence signals under complex ambient light interference. This significantly improves detection efficiency, data accuracy, and environmental adaptability for in situ quasi-3D fluorescence spectroscopy. Experimental validation demonstrates the ability of the method to rapidly and accurately acquire high-resolution fluorescence spectra at different distances, significantly enhancing the signal-to-noise ratio (SNR). Compared to traditional hardware lock-in techniques, the software-based approach offers greater flexibility and scalability in high-throughput, multi-parameter weak signal detection, with broad potential applications in phytoplankton research and harmful algal bloom monitoring.

Our method utilizes a custom Simulink program to implement software lock-in amplification, enhancing fluorescence signals collected by the spectrometer while effectively mitigating environmental light interference, such as sunlight. The process begins with heterodyne modulation, where multi-wavelength excitation spectra are synchronized with reference signals using sine or cosine waves of different frequencies, mapping the spectral signals to corresponding frequency components. The lock-in amplification module compares the phase of each signal with the reference, amplifying target signals while suppressing background noise and high-frequency components. A low-pass filter is applied to remove residual high-frequency noise and elastic light signals from other modulation frequencies, further improving signal accuracy. Experimental tests validate the method’s effectiveness in eliminating environmental light interference, demonstrating its ability to extract weak fluorescence signals under complex ambient conditions. Tests conducted at distances of 10 m and 5 m reveal significant signal enhancement, with real-time parameter adjustments available to optimize lock-in amplification settings based on experimental conditions. Simulink’s graphical interface facilitates on-site adjustments, ensuring stable, high-precision fluorescence signal detection across diverse environments. This method not only enhances signal accuracy and stability but also provides robust resistance to interference, showcasing broad application potential.

Experimental results confirm the effectiveness of the software lock-in amplification method in improving SNR and eliminating environmental light interference. Initial tests under sunlight conditions reveal that fluorescence signals are heavily obscured, resulting in low SNR and limited extractable information (Fig. 6). After applying the software lock-in amplification technique, fluorescence signals are extracted (Figs. 7 and 8). At a distance of 10 m, the SNR improves from -9.7 to 15.56 dB, while at 5 m, the SNR increases to 25.38 dB. These results highlight the significant improvements in detection accuracy and stability of fluorescence signals by the software lock-in amplification method, even under complex ambient light conditions. In addition, the normalized fluorescence spectra of four distinct phytoplankton are illustrated (Fig. 9). By extracting and analyzing fluorescence signals from various algal samples, this method facilitates the efficient classification of phytoplankton and extraction of relevant biological information. The experimental findings demonstrate the method’s strong adaptability for phytoplankton detection, enabling precise and efficient fluorescence spectrum measurements across diverse experimental conditions.

In this paper, we propose an innovative parallel acquisition method for quasi-3D fluorescence spectra using software lock-in amplification. By employing heterodyne modulation and combining multi-wavelength excitation sources with lock-in amplification technique, this method significantly improves the SNR in complex environments and successfully isolates characteristic fluorescence signals, such as chlorophyll a. Through frequency division multiplexing and algorithm optimization, the approach enhances both the speed and stability of data collection. Compared to traditional hardware-based methods, the software solution offers notable advantages in flexibility, scalability, and cost-effectiveness, particularly for high-throughput spectral data processing. Experimental results demonstrate the effectiveness of the proposed method in suppressing background noise and accurately differentiating fluorescence features of various algae species, providing a reliable foundation for algae classification and ecological monitoring. Further work will explore the integration of advanced classification algorithms and hardware optimization to further enhance practical applications and accuracy across diverse fields.