Published 2024-02-28
Keywords
- Microplastics; Marine pollution; Hyperspectral; Support vector machine.
How to Cite
This work is licensed under a Creative Commons Attribution 4.0 International License.
Abstract
Addressing the rising concern of microplastic pollution in soil, this study proposes employing hyperspectral imaging technology for efficient detection. Utilizing supervised classification algorithms, including Support Vector Machine (SVM), Mahalanobis Distance (MD), and Maximum Likelihood (ML), microplastic pollutants in soil are directly identified and classified. Experiments conducted within a wavelength range of 400-1000 nm reveal SVM as the most suitable algorithm, achieving an average identification accuracy of 84% for white polyethylene (PE) microplastics within the 1-5 mm particle size range.