We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Deep Learning-BasedShape Classification for Hyperspectral-ImagedMicroplastics
Summary
Researchers tested nine deep learning architectures for automating shape classification of microplastic particles in hyperspectral images, comparing performance across original and augmented datasets. The best-performing architectures achieved high accuracy, offering a faster and more consistent alternative to manual expert classification.
The shape of microplastics (MPs) matters. Yet, the expert-based shape classification is labor intensive, time consuming, and susceptible to human biases. In this study, we investigated deep learning-based approaches for automating shape classifications of MP hyperspectral images, thus achieving a faster and more accurate classification procedure. Here, nine deep learning architectures (NN1.1, NN 1.2, CNN 1.1, CNN 1.2, CNN 1.3, VGG16, ResNet50, ResNet50 V2, and MobileNet) were tested and further compared in terms of their performance discrepancies across four data sets (original, augmented original, refined, and augmented refined data sets). Our sample images comprise the hyperspectral images of 11,042 environmental MP, (particle sizes down to 10 μm) analyzed with micro-Fourier transform infrared spectroscopy, covering seven environmental matrices (wastewater influent, effluent, sludge, marine water, stormwater, sediments from stormwater ponds, and indoor air). Nine shape categories, including fiber, rod, ellipse, oval, sphere, quadrilateral, triangle, free-form, and unidentifiable were applied as reference shapes. Based on the comparison test, three main findings are outlined: (a) Model architecture influences MP shape classification significantly, where CNNs outperform NNs and transfer learning-based models outperform nontransfer learning-based models. Notably, MobileNet achieves the highest accuracy of 0.93 and 1.00 in validation/test data sets, respectively. (b) Data quality matters for shape classification, where complex models demonstrate robust performance across data sets while simple models are more sensitive to data quality changes. (c) In contrast to manual assessment, the deep learning approach has achieved an automated shape classification process for hyperspectral images, which reduces the consumption of labor and time while increasing efficiency significantly. Yet, challenges and potential remain, particularly regarding model architecture and data quality, highlighting the need for robust designs and complementing high-quality data sets for optimal classification.
Sign in to start a discussion.
More Papers Like This
Deep Learning-Based Shape Classification for Hyperspectral-Imaged Microplastics
Researchers tested nine deep learning architectures for automating the shape classification of microplastic particles in hyperspectral images, comparing performance on original and augmented datasets. The best models achieved high classification accuracy, offering a faster and more consistent alternative to labour-intensive manual identification.
Self-Supervised Hierarchical Dilated Transformer Network for Hyperspectral Soil Microplastic Identification and Detection
A self-supervised hierarchical dilated transformer neural network was developed for automated classification of microplastic images, achieving high accuracy across multiple polymer types. The deep learning approach reduces the labor-intensive manual identification step in microplastic analysis workflows.
Application of hyperspectral imaging and machine learning for the automatic identification of microplastics on sandy beaches
Hyperspectral imaging combined with machine learning was applied to identify and classify microplastics on sandy beach surfaces, offering a faster and more scalable alternative to conventional spectroscopic analysis for large-area environmental monitoring.
Development of robust models for rapid classification of microplastic polymer types based on near infrared hyperspectral images
Researchers used near-infrared hyperspectral imaging combined with machine learning to classify nine types of microplastic particles, finding reliable results even for small particles on wet filters. This method could enable faster, automated identification of diverse microplastic types in environmental water samples.
Microplastic Spectral Classification Using Deep Learning with Denoising and Dimensionality Reduction
Researchers developed a deep learning approach for microplastic spectral classification that incorporates denoising and dimensionality reduction steps, improving the accuracy of identifying and classifying microplastic polymer types from spectral data in marine ecosystems.