A Feature Domain Space Transfer Method for Improving Identification of Maize Haploid Seed Based on Near-infrared Spectroscopy

Lin, Jianchu; Zhang, Liping; Li, Jinlong; Zhou, Wen

doi:10.1109/hpbdis.2019.8735481

Cited by 1 publication

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For this task, domain adaptation can be used to train an ML model on the spectra acquired under stationary conditions (the source domain), akin to a manufacturer collecting offline data in a laboratory, and aid transfer of these models to the moving conditions on a conveyor line (the task domain). Matrix or kernel-based methods have previously been used for domain adaptation of 1D NIR spectra to find projections between domains [ 21 , 22 ] or to also extract discriminative features to be used for a main learning task [ 23 , 24 , 25 , 26 ]. Ensemble methods that maximize model diversity and prediction similarity in the new domain data have also been used to increase the probability that transferable correlations are used [ 27 ].…”

Section: Introductionmentioning

confidence: 99%

Domain Adaptation for In-Line Allergen Classification of Agri-Food Powders Using Near-Infrared Spectroscopy

Bowler

Ozturk

Rady

et al. 2022

Sensors

View full text Add to dashboard Cite

The addition of incorrect agri-food powders to a production line due to human error is a large safety concern in food and drink manufacturing, owing to incorporation of allergens in the final product. This work combines near-infrared spectroscopy with machine-learning models for early detection of this problem. Specifically, domain adaptation is used to transfer models from spectra acquired under stationary conditions to moving samples, thereby minimizing the volume of labelled data required to collect on a production line. Two deep-learning domain-adaptation methodologies are used: domain-adversarial neural networks and semisupervised generative adversarial neural networks. Overall, accuracy of up to 96.0% was achieved using no labelled data from the target domain moving spectra, and up to 99.68% was achieved when incorporating a single labelled data instance for each material into model training. Using both domain-adaptation methodologies together achieved the highest prediction accuracies on average, as did combining measurements from two near-infrared spectroscopy sensors with different wavelength ranges. Ensemble methods were used to further increase model accuracy and provide quantification of model uncertainty, and a feature-permutation method was used for global interpretability of the models.

show abstract