We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Laser-induced breakdown spectroscopy with neural network approach for plastic identification and classification in waste management
Summary
Researchers applied laser-induced breakdown spectroscopy combined with neural network algorithms to identify and classify different plastic types, addressing the need for rapid and accurate plastic sorting in recycling chains. The system demonstrated high classification accuracy for common polymer types based on their elemental emission spectra.
The threats to the environment and humans are increasing every day due to the use of modern plastics and their improper disposal approaches. Researchers pay more attention to reducing plastic waste through recycling so that it can be used as a raw material. In the recycling chain, grading or identifying different types of plastic is essential. For this, Lase Induced Breakdown Spectroscopy (LIBS) has been established. LIBS is an effective investigation tool that analyzes plastics in a qualitative and quantitative manner. Spectral analysis of different kinds of plastics is performed from the plasma emission obtained from LIBS. In this research work different types of plastic samples are identified using LIBS and classified using back propagation neural network algorithm (BPNN). The research aimed to attain a simple application to detect plastic polymers compared to existing approaches. To validate the better results proposed model performances are compared with existing kNN, SIMCA and ANN based classification models.
Sign in to start a discussion.
More Papers Like This
Identification of 20 polymer types by means of laser-induced breakdown spectroscopy (LIBS) and chemometrics
Researchers developed a laser-based identification technique that can distinguish among 20 different types of plastic using chemical analysis and machine learning, even in colored or additive-containing samples — a higher number than any previously published method. Rapid and reliable plastic identification is a critical step for improving plastic waste sorting and understanding the composition of environmental microplastic pollution.
Identification of marine microplastics by laser-induced fluorescence spectroscopy: 1-Dimensional convolutional neural network and continuous convolutional model
Researchers investigated using laser-induced fluorescence spectroscopy combined with deep learning models to identify six types of marine microplastics. A continuous convolution neural network model achieved 99.5% classification accuracy, outperforming a standard 1D convolutional network at 97.5%. The approach offers a faster and less expensive alternative to traditional FTIR and Raman spectroscopy methods for microplastic identification.
Online in situ detection of atmospheric microplastics based on laser-induced breakdown spectroscopy
Researchers developed a laser-based detection system combined with machine learning that can identify and classify different types of microplastics in the air in real time. The system achieved high accuracy in distinguishing between common plastic types like polyethylene, polystyrene, and PVC. Better tools for monitoring airborne microplastics are important because people inhale these particles daily, and understanding what types are present in the air is the first step toward assessing respiratory health risks.
Rapid Identification of Beached Marine Plastics Pellets Using Laser-Induced Breakdown Spectroscopy: A Promising Tool for the Quantification of Coastal Pollution
Researchers applied laser-induced breakdown spectroscopy combined with chemometric analysis to rapidly identify and classify beached plastic pellets by polymer type, achieving over 80% accuracy and demonstrating its potential as a fast, field-deployable tool for coastal pollution monitoring.
Deep learning for chemometric analysis of plastic spectral data from infrared and Raman databases
A novel deep learning architecture called PolymerSpectraDecisionNet was trained to identify common recyclable plastics from infrared and Raman spectral databases. The model outperformed conventional chemometric methods for polymer classification and was designed to handle real-world spectral variability relevant to the plastics recycling industry.