Data_PE_3_4

Description

This dataset is a subset of the “2T2D spectroscopy and deep learning for soil MPs quantification” data, comprising 2T2D spectra of soil and microplastics (MPs) mixtures.

Steps to reproduce

First, the infrared (IR) spectra of MPs mixtures were acquired using a Fourier transform infrared spectrometer. Then, the IR spectra were subjected to data augmentation, baseline correction, Savitzky-Golay filtering, standard normal variate and multiplicative scatter correction processing. Subsequently, they were transformed into 2T2D spectra (auto-synchronous, auto-asynchronous, sy-asynchronous, cross-synchronous and cross-asynchronous spectra). Finally, the combination of 2T2D spectroscopy and a deep residual shrinkage network embedded with channel attention and strip attention (DRSN-SpCA) was employed for the quantitative analysis of PE in soil.

Institutions

• Hefei University of Technology

  Anhui, Hefei

Categories

Licence