We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Detection of Microplastics in Freshwater Sediments Based on Raman Spectroscopy and Convolutional Neural Networks
Summary
Researchers developed a method combining Raman spectroscopy and convolutional neural networks to detect and classify microplastics in complex freshwater sediment samples, training the CNN on mixed spectra from extracted sediment fractions to improve detection accuracy.
Monitoring of microplastics (MPs) in aquatic environments is essential for the prevention and control of MP pollution. In aquatic environments, MPs tend to sink into sediments, resulting in a much higher concentration of MPs in sediments than in water. However, the composition of sediments is very complex and difficult to analyze. In this article, a detection method for MPs in freshwater sediments based on Raman spectroscopy and convolutional neural networks (CNNs) has been developed. Density separation and vacuum-assisted filtration were employed to separate MPs from sediments. The CNN model was trained using Raman spectra of MPs and organic mixtures extracted from freshwater sediments and tested using Raman spectra of MPs and sediment mixtures. The identification accuracy for MPs using CNN reached as high as 94.27%, outperforming other machine learning models such as the support vector machine (SVM) and random forest (RF). The proposed method was applied to the detection of MPs in the sediments of Changdang Lake, revealing that MPs in Changdang Lake primarily originate from plastic waste generated by human activities around the lake rather than from river inflows.