Fluorescence-assisted image analysis of freshwater microalgae

Area of Science:

• Environmental science
• Microbiology
• Image processing

Background:

• Automated analysis of microalgae in sediment and water samples is challenging due to complex scenes and interfering objects.
• Traditional methods struggle with accuracy and speed when dealing with natural populations and diverse sample matrices.

Purpose of the Study:

• To develop and validate a novel fluorescence-based image processing system for enhanced microalgae detection and classification.
• To improve the accuracy and efficiency of analyzing microalgae in challenging environmental samples like sediment and water.

Main Methods:

• Incorporated fluorescence excitation into a microalgae image processing system.
• Quantitatively measured 120 object characteristics detected via fluorescence, using an optimized subset for automated analysis.
• Employed template matching and automated seeded region growing (SRG) to resolve object drift issues between fluorescence and greyscale images.

Main Results:

• The system successfully detected and analyzed microalgae in complex sediment and water samples.
• Achieved high classification accuracy (>97%) for two major microalgae genera: Anabaena spp. and Microcystis spp.
• Demonstrated significant improvements in processing speed and classification accuracy compared to non-fluorescence methods.

Conclusions:

• Fluorescence excitation is crucial for overcoming limitations in automated microalgae analysis in complex environmental samples.
• The developed system offers a user-friendly and highly accurate solution for microalgae species classification.
• This approach has significant implications for ecological monitoring and water quality assessment.