Work

Underwater Phaeocystis globosa monitoring based on ACE–DCP enhancement and YOLOv8 detection

Year: 2025

Type: article

Abstract: Harmful algal blooms have emerged as a formidable menace to both human society and the ecological environment. Specifically, the occurrence of Phaeocystis globosa blooms poses a potential hazard to the safety of coastal nuclear power cooling systems. However, current ecological monitoring techniques fail to dynamically detect the densities of solitary cells of Phaeocystis globosa prior to the blooms, thus hindering timely interventions. This study proposes a framework for harmful algae monitoring by integrating underwater microscopic imaging, image processing, and object detection. Flume experiments were conducted using Phaeocystis globosa as the case study for monitoring object. The results indicate that the proposed framework exhibits favorable performance in recognizing different types of algae, particularly in distinguishing between Phaeocystis globosa and Chlorella. Despite their similar morphology observed from the underwater imaging device under dark field illumination, the false detection rate between Phaeocystis globosa and Chlorella is approaches 0 % when using the YOLOv8 object detection model. Adaptive Contrast Enhancement (ACE) amplifies the color discrepancies among algae and eliminates virtual focus interference, thus improving precision of algae classification. Subsequently, Dark Channel Prior (DCP) reduces the noise caused by image scattering and limits missed detection. Consequently, the accuracy of Phaeocystes globosa recognition using the YOLOv8 model in increased from 74 % to 91 %. This study presents an effective solution for in-situ monitoring of specific harmful algae, which has the potential to enhance the capabilities for dynamic detection and early warning of Phaeocystis Globosa blooms.

Source: Applied Optics

DOI: 10.1364/ao.549613