Related Experiment Video
Updated: Jan 13, 2026
08:27
Sampling and Identification of Microplastics in Groundwater
Published on: November 7, 2025
903
Identification and velocity measurement of microplastics based on machine learning
1School of Marine Engineering Equipment, Zhejiang Ocean University, Zhoushan City, Zhejiang Province, 316022, China.
Water Research
|October 28, 2025
Summary
This study introduces a novel machine learning framework for simultaneously tracking multiple microplastics (MPs) and measuring their settling velocities in aquatic environments. This advanced method enhances understanding of MP migration and particle interactions.
Area of Science:
- Environmental Science
- Fluid Dynamics
- Artificial Intelligence
Background:
- Microplastic (MP) settling velocity is crucial for aquatic transport.
- Current methods struggle with simultaneous multi-particle tracking and interactions.
- Understanding MP dynamics requires advanced tracking techniques.
Purpose of the Study:
- To develop a novel machine learning framework for simultaneous multi-MP tracking.
- To accurately measure individual terminal settling velocities of MPs.
- To investigate hydrodynamic particle-particle interactions during sedimentation.
Main Methods:
- Integration of an enhanced YOLOv5-CA object detection model with DeepSort tracking.
- Utilizing a square glass sedimentary column for MP sedimentation experiments.
- High-throughput simultaneous tracking of multiple microplastic particles.
Main Results:
- Achieved a 99% tracking success rate and 85.3% mean accuracy.
- Demonstrated a maximum error of 1.7% compared to conventional methods.
- Enabled the study of hydrodynamic particle-particle interactions during sedimentation.
Conclusions:
- The developed framework enables efficient, simultaneous multi-particle tracking of MPs.
- Accurate measurement of MP settling velocities is now feasible.
- This facilitates systematic investigation of particle-particle interactions in MP transport research.
