Microalgal density assessment based on quantum-dot light-emitting diodes and intelligent image edge detection

This study proposes an approach involving image capture, recognition, and processing using quantum dots for light conversion, which emit blue, green, orange, and red light. Microalgae species of Nannochloropsis sp. and Chaetoceros sp. are selected for observation. In addition to color differences and brightness distribution, an optimized composite detecting indicator (O_CDI) is introduced, which combines six conventional edge detection indicators with adjustable weight coefficients. The performance obtained with the quantum-dot illumination system is compared with results obtained under non-specific lighting conditions by evaluating image edge characteristics, average brightness, color differences, and light intensity measurements. When evaluating O_CDI under different colors, the proposed method achieves an accuracy (η) of 0.99 and a coefficient of determination (R²) of 0.99, as compared to conventional manual counting method. The proposed microalgal density assessment method, characterized by high accuracy, flexibility, and environmental friendliness, demonstrates potential applicability in smart marine agriculture and digital marine monitoring.