The Identification of Spherical Engineered Microplastics and Microalgae by Micro-hyperspectral Imaging.
Bull Environ Contam Toxicol
; 107(4): 764-769, 2021 Oct.
Article
em En
| MEDLINE
| ID: mdl-33599786
ABSTRACT
Based on the micro-hyperspectral imaging technique, spherical engineered microplastic (polyethylene, 10-45 µm) and microalgae (Isochrysis galbana) (4-7 µm) were identified. In transmittance mode of MHSI, micro image cubes from 400 to 1000 nm were obtained from slides containing MP and MA in thin seawater. Classifiers like Support Vector Machine (SVM(Radial Basis Function (RBF))), Least Squares Support Vector Machine (LSSVM(RBF)), k-Nearest Neighbors, etc. were adopted and compared to classify MP and MA. In order to expand the imaging range of micro imaging, image stitching technology was adopted. In allusion to the stitched image cube, SVM(RBF) is suggested for the identification of MA and MP, with recall and precision > 0.86. The above results demonstrate that the MHSI is a promising technique, which can detect MPs with particle size Limit of Detection of 10-45 µm, and it is potential to further expand this LOD.
Palavras-chave
Texto completo:
1
Coleções:
01-internacional
Base de dados:
MEDLINE
Assunto principal:
Haptófitas
/
Microalgas
Tipo de estudo:
Diagnostic_studies
Idioma:
En
Revista:
Bull Environ Contam Toxicol
Ano de publicação:
2021
Tipo de documento:
Article
País de afiliação:
China