We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Automatic microplastic classification using dual-modality spectral and image data for enhanced accuracy
AI summary Read the abstract
A dual-modality classification system combining FTIR spectral data and microscope images achieved 99% accuracy in automatically identifying five common microplastic polymer types. The study deployed a web application (MPsSpecClassify) that enables researchers to efficiently classify microplastics, addressing the time-consuming and error-prone nature of manual spectral analysis.
The development of an automatic microplastic (MPs) classification system using spectra is crucial due to the time-consuming and error-prone nature of analyzing individual spectra, especially with a large quantity of MPs. This study presents a classification system using a dual-modality dataset from micro-Fourier Transform Infrared Spectroscopy (μFTIR) for five common polymer types: polypropylene, polystyrene, polyethylene terephthalate, polyethylene, and polyamide. A comparison of machine learning models, including Decision Tree (DT), Extremely Randomized Trees (ET), Support Vector Classifier (SVC), and Multiclass Logistic Regression (LR), is conducted using features extracted by AlexNet, ResNet18, and Vision Transformer (ViT). Notably, the AlexNet with Logistic Regression (AlexNet-LR) model demonstrated exceptional performance, achieving a validation accuracy of 99.03 % and nearly perfect test scores of 99.99 %. However, ResNet18-LR was selected for web deployment due to its shorter training and inference times compared to AlexNet-LR, while still achieving 99 % validation and test accuracy. This highlights the effectiveness of using a dual-modality dataset for precise microplastic classification. MPsSpecClassify, a web-based application, was developed to enable users to efficiently identify MPs and improve microplastic pollution management.
More Papers Like This
A Comparative Study of Machine Learning and Deep Learning Models for Microplastic Classification using FTIR Spectra
AI summary Read the abstract
Researchers compared machine learning and deep learning models for classifying microplastics using FTIR spectra, evaluating multiple algorithmic approaches against standardised spectral datasets. The study assessed classification accuracy and computational efficiency, identifying which model architectures best discriminate between polymer types in environmental microplastic samples.
Development of a rapid detection protocol for microplastics using reflectance-FTIR spectroscopic imaging and multivariate classification
AI summary Read the abstract
Reflectance-FTIR spectroscopy was evaluated as a faster and more automated detection method for microplastics in environmental samples, with results showing strong potential for high-throughput screening. The method could reduce the time and cost of routine microplastic monitoring programs.
Robust Automatic Identification of Microplastics in Environmental Samples Using FTIR Microscopy
AI summary Read the abstract
Researchers developed a robust automated method for identifying microplastics in environmental samples using FTIR microscopy combined with machine learning-based spectral matching, improving the consistency and efficiency of microplastic identification compared to manual evaluation.
Classifying polymers with mid-IR spectra and machine learning: From monitoring to detection
AI summary Read the abstract
Researchers applied machine learning to mid-infrared spectra to automatically classify different types of plastic polymers found in the environment. Accurate polymer identification is essential for microplastic research, and this automated approach could improve monitoring efficiency and data consistency across studies.
Development of a machine‐learning model for microplastic analysis in an FT‐IR microscopy image
AI summary Read the abstract
Researchers developed a machine-learning model using a 1D convolutional neural network to classify FT-IR microscopy spectra of microplastics into 16 polymer types. The model addresses inaccuracies caused by secondary materials on real environmental samples, improving the speed and reliability of automated microplastic identification.
Research digests by email
When a large batch of papers lands in the Atlas, we read through it and send a short write-up of what stood out.