Analyzing Near-Infrared Hyperspectral Imaging for Protein Content Regression and Grain Variety Classification Using Bulk References and Varying Grain-to-Background Ratios

Publikation: Bog/antologi/afhandling/rapport › Rapport › Forskning

Standard

Analyzing Near-Infrared Hyperspectral Imaging for Protein Content Regression and Grain Variety Classification Using Bulk References and Varying Grain-to-Background Ratios. / Engstrøm, Ole-Christian Galbo; Dreier, Erik Schou; Jespersen, Birthe P Møller; Steenstrup Pedersen, Kim.

arxiv.org, 2023. 21 s.

Publikation: Bog/antologi/afhandling/rapport › Rapport › Forskning

Harvard

Engstrøm, O-CG, Dreier, ES, Jespersen, BPM & Steenstrup Pedersen, K 2023, Analyzing Near-Infrared Hyperspectral Imaging for Protein Content Regression and Grain Variety Classification Using Bulk References and Varying Grain-to-Background Ratios. arxiv.org.

APA

Engstrøm, O-C. G., Dreier, E. S., Jespersen, B. P. M., & Steenstrup Pedersen, K. (2023). Analyzing Near-Infrared Hyperspectral Imaging for Protein Content Regression and Grain Variety Classification Using Bulk References and Varying Grain-to-Background Ratios. arxiv.org.

Vancouver

Engstrøm O-CG, Dreier ES, Jespersen BPM, Steenstrup Pedersen K. Analyzing Near-Infrared Hyperspectral Imaging for Protein Content Regression and Grain Variety Classification Using Bulk References and Varying Grain-to-Background Ratios. arxiv.org, 2023. 21 s.

Author

Engstrøm, Ole-Christian Galbo ; Dreier, Erik Schou ; Jespersen, Birthe P Møller ; Steenstrup Pedersen, Kim. / Analyzing Near-Infrared Hyperspectral Imaging for Protein Content Regression and Grain Variety Classification Using Bulk References and Varying Grain-to-Background Ratios. arxiv.org, 2023. 21 s.

Bibtex

@book{96253329d6c74db79c5dac401bf7727b,
title = "Analyzing Near-Infrared Hyperspectral Imaging for Protein Content Regression and Grain Variety Classification Using Bulk References and Varying Grain-to-Background Ratios",
abstract = "Based on previous work, we assess the use of NIR-HSI images for calibrating models on two datasets, focusing on protein content regression and grain variety classification. Limited reference data for protein content is expanded by subsampling and associating it with the bulk sample. However, this method introduces significant biases due to skewed leptokurtic prediction distributions, affecting both PLS-R and deep CNN models. We propose adjustments to mitigate these biases, improving mean protein reference predictions. Additionally, we investigate the impact of grain-to-background ratios on both tasks. Higher ratios yield more accurate predictions, but including lower-ratio images in calibration enhances model robustness for such scenarios.",
author = "Engstr{\o}m, {Ole-Christian Galbo} and Dreier, {Erik Schou} and Jespersen, {Birthe P M{\o}ller} and {Steenstrup Pedersen}, Kim",
year = "2023",
language = "Dansk",
publisher = "arxiv.org",

}

RIS

TY - RPRT

T1 - Analyzing Near-Infrared Hyperspectral Imaging for Protein Content Regression and Grain Variety Classification Using Bulk References and Varying Grain-to-Background Ratios

AU - Engstrøm, Ole-Christian Galbo

AU - Dreier, Erik Schou

AU - Jespersen, Birthe P Møller

AU - Steenstrup Pedersen, Kim

PY - 2023

Y1 - 2023

N2 - Based on previous work, we assess the use of NIR-HSI images for calibrating models on two datasets, focusing on protein content regression and grain variety classification. Limited reference data for protein content is expanded by subsampling and associating it with the bulk sample. However, this method introduces significant biases due to skewed leptokurtic prediction distributions, affecting both PLS-R and deep CNN models. We propose adjustments to mitigate these biases, improving mean protein reference predictions. Additionally, we investigate the impact of grain-to-background ratios on both tasks. Higher ratios yield more accurate predictions, but including lower-ratio images in calibration enhances model robustness for such scenarios.

AB - Based on previous work, we assess the use of NIR-HSI images for calibrating models on two datasets, focusing on protein content regression and grain variety classification. Limited reference data for protein content is expanded by subsampling and associating it with the bulk sample. However, this method introduces significant biases due to skewed leptokurtic prediction distributions, affecting both PLS-R and deep CNN models. We propose adjustments to mitigate these biases, improving mean protein reference predictions. Additionally, we investigate the impact of grain-to-background ratios on both tasks. Higher ratios yield more accurate predictions, but including lower-ratio images in calibration enhances model robustness for such scenarios.

M3 - Rapport

BT - Analyzing Near-Infrared Hyperspectral Imaging for Protein Content Regression and Grain Variety Classification Using Bulk References and Varying Grain-to-Background Ratios

PB - arxiv.org

ER -

ID: 375984534