Year: 2025
Type: conference-paper
Abstract: Accurate object detection is critical in marine and ecological research, where microscopic imaging provides insights into organism health and environmental conditions. This paper presents a comparative study of object detection using YOLOv5 and YOLOv8 for three distinct classes: artemia, cyst, and excrement. Our methodology involves bounding box annotation, contour extraction, and fluorescence-based intensity measurements to assess nanoparticle uptake in Artemia. In this comparative study, we analyze the performance of these models in terms of accuracy, precision, recall, etc. where YOLOv5 exhibits higher detection rates for Artemia and excrement, demonstrating remarkable robustness in identifying these classes despite challenging image conditions. However, YOLOv8 outperforms YOLOv5 for cyst detection, suggesting it may better handle this specific class. This suggests that YOLOv5 offers greater versatility and adaptability in detection tasks while YOLOv8 may struggle in difficult situations and may need further fine-tuning or specialized training to enhance its performance. The results show insights into the suitability of YOLOv5 and YOLOv8 for detecting objects in challenging marine environments, with implications for applications such as ecological research.
Source: Proceedings of IEEE Southeastcon
Institution: Florida International University
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
SDG: Life below water
Open Access status: green
Funder: National Science Foundation
OpenAlex ID: W4409797016