Detection Of Environmental Microplastics Based On Hyperspectral Imaging Technology

Microplastics(MPs)pollution in marine systems and terrestrial soil ecosystems is attracting worldwide attentions.Research and reports about marine and terrestrial MPs pollution have been substantially intensified in recent decade.At present,MPs particles can be accurately detected by Raman or Fourier transform infrared spectroscopy(FT-IR)spectroscopy,which requires individual particle analysis,indicating the suspected particles need to be identified visually and operated manually one by one.To reach a statistically reliable statement about the occurrence of MPs present in the environment,a large number of particles must be analyzed,which would be time consuming.In view of this,there is an urgent need to develop a new technology for quickly and accurately determining and mapping the distribution of MPs in seawater and soil.Hyperspectral imaging technology has been investigated as a possible way to detect MPs contamination in soil directly and efficiently in this study.In the study,hyperspectral system with wavelength range between 400 and 1000 nm was obtained hyperspectral images of MPs soil samples containing different materials.Supervised classification algorithms such as support vector machine(SVM),mahalanobis distance(MD)and maximum likelihood(ML)algorithms were then used to identify MPs from the other materials in hyperspectral images.To investigate the effect of particle size and color,white polyethylene(PE)and black PE particles extracted from soil with two different particle size ranges(1-5 mm and 0.5-1 mm)were studied in this work.The results showed that SVM was the most applicable method for detecting white PE MPs in soil,with the precision of 84% and 77% for PE particles in size ranges of 1-5 mm and 0.5-1 mm respectively.Finally,six kinds of household polymers including drink bottle,bottle cap,rubber,packing bag,clothes hanger and plastic clip were used to validate the developed SVM method,and the classification precision of polymers were obtained from 79%-100% and 86%-99% for MPs particle 1-5 mm and 0.5-1 mm,respectively.The hyperspectral imaging system with a wavelength range of 900-1700 nm was used in the study of charactering MPs in seawater.This work aimed to identify MPs from the hyperspectral image using SVM algorithm,which presents a good performance for analyzing nonlinear and highdimensional data and is insensitive to the Hughes effect.Hyperspectral images of MP samples of seawater and filtered seawater were collected.The SVM algorithm was then used to process hyperspectral image,quantification and identification of MPs in seawater and filtered seawater samples,to study the effect of seawater on the hyperspectral identification of MPs.The results showed that the detection limit of the MPs in the hyperspectral imaging system was increased to 0.2 mm when the seawater in samples was removed.Other factors which may affect the accuracy of SVM model were investigated,including organic particles,polymer types and particle sizes.SVM model yielded an accuracy of above 90% for the identification of common household polymers and eight pure polymer MPs in the samples,and it presented a highly robust for detecting MPs in a wide range of types,particle sizes and colors.The results indicate that hyperspectral imaging technology is a potential MPs detection method,and the MPs particle in the soil surface and seawater could be directly determined and visualized.