Spectroscopic Identification of Environmental Microplastics

Spectroscopic technology is widely used in identifying the categories of microplastics (MPs) for its non-destructive, rapid, and without pretreatment characters. Recognition spectral category often conducted by matching with reference library, this works well a known material but fails to blindly identify unknown source environmental MPs. In work, robust classifier was proposed differentiate chemical types MPs samples, recognition rate higher than 0.97. This innovatively an adaptive estimator developed k-nearest neighbor (kNN) model as hard threshold classify MPs, thus interference distortions diversity effectively eliminated. method increases ability interpret spectra realistic samples.