We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
An automated deep learning pipeline based on advanced optimisations for leveraging spectral classification modelling
Summary
Researchers developed an automated system that uses advanced optimization methods to automatically design and tune deep learning models for analyzing spectral data — light-based fingerprints used to identify materials. Applied to wheat variety classification, the system achieved 94.9% accuracy with a simpler model architecture than previously reported methods.
In deep learning (DL) modelling for spectral data, a major challenge is related to the choice of DL network architecture and the selection of the best hyperparameters. Often, slight changes to the neural architecture or its hyperparameter can have a direct influence on the model's performance, making its robustness questionable. To deal with it, this study presents an automated deep learning modelling based on advanced optimisation techniques involving Hyperband and Bayesian optimisation, to automatically find optimal neural architecture and its hyperparameters to reach robust DL models. The optimisation requires a base neural architecture to be initialized, however, later it automatically adjusts the neural architecture and the hyperparameters to reach the optimal model. Furthermore, to support the interpretation of the DL models, a wavelength weighing schema based on gradient-weighted class activation mapping (Grad-CAM) was implemented. The potential of the approach was showed on a real case of wheat variety classification with near-infrared spectral data. The performance of the classification was compared with that previously reported on the same dataset with different DL and chemometric approaches. The results showed that with the proposed approach a classification accuracy of 94.9% was reached, which was better than the best reported accuracy on the same data set i.e., 93%. Furthermore, the better performance was obtained with a simpler neural architecture compared to what was used in earlier studies. The automated deep learning based on advanced optimisation can support DL modelling of spectral data.
Sign in to start a discussion.
More Papers Like This
Optimized recognition of microplastic ATR-FTIR spectra with deep learning
Researchers developed an optimized deep learning method for identifying microplastics from ATR-FTIR spectra, improving classification accuracy for weathered and environmentally contaminated MP samples that challenge standard spectral library matching approaches.
Deep learning-powered efficient characterization and quantification of microplastics
Researchers developed an artificial intelligence framework that uses deep learning to automatically identify and quantify microplastics from infrared spectra and visual images. The system achieved high accuracy in classifying plastic types and counting particles, dramatically reducing the time needed compared to manual analysis. This tool could make large-scale microplastic monitoring faster and more consistent across different research laboratories.
Detection of seed purity of hybrid wheat using reflectance and transmittance hyperspectral imaging technology
Researchers developed a non-destructive hyperspectral imaging method to distinguish hybrid wheat seeds from parent seeds with up to 98% accuracy, offering a faster and more reliable alternative to traditional seed purity testing methods for the agriculture industry.
PlasticNet: Deep Learning for Automatic Microplastic Recognition via FT-IR Spectroscopy
Researchers developed PlasticNet, a deep learning algorithm that automatically identifies microplastic types from infrared spectral data, outperforming conventional library matching approaches. Automating microplastic identification could dramatically speed up the analysis of environmental samples and reduce human error.
Identification of Polymers with a Small Data Set of Mid-infrared Spectra: A Comparison between Machine Learning and Deep Learning Models
Researchers compared multiple machine learning and deep learning models for identifying polymer types from mid-infrared spectral data using a small reference dataset, finding that certain deep learning architectures outperformed traditional methods even with limited training examples, supporting automated microplastic identification.