Aarhus University Seal / Aarhus Universitets segl

Birte Boelt

Optimal sample size for predicting viability of cabbage and radish seeds based on near infrared spectra of single seeds

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Standard

Optimal sample size for predicting viability of cabbage and radish seeds based on near infrared spectra of single seeds. / Shetty, Nisha; Min, Tai-Gi; Gislum, René; Olesen, Merete Halkjær; Boelt, Birte.

In: Journal of Near Infrared Spectroscopy, Vol. 19, No. 6, 2011, p. 451-461.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Harvard

APA

CBE

MLA

Vancouver

Author

Shetty, Nisha ; Min, Tai-Gi ; Gislum, René ; Olesen, Merete Halkjær ; Boelt, Birte. / Optimal sample size for predicting viability of cabbage and radish seeds based on near infrared spectra of single seeds. In: Journal of Near Infrared Spectroscopy. 2011 ; Vol. 19, No. 6. pp. 451-461.

Bibtex

@article{e403991308c8478aaa5e71001cffdecd,
title = "Optimal sample size for predicting viability of cabbage and radish seeds based on near infrared spectra of single seeds",
abstract = "The effects of the number of seeds in a training sample set on the ability to predict the viability of cabbage or radish seeds are presented and discussed. The supervised classification method extended canonical variates analysis (ECVA) was used to develop a classification model. Calibration sub-sets of different sizes were chosen randomly with several iterations and using the spectral-based sample selection algorithms DUPLEX and CADEX. An independent test set was used to validate the developed classification models. The results showed that 200 seeds were optimal in a calibration set for both cabbage and radish data. The misclassification rates at optimal sample size were 8%, 6% and 7% for cabbage and 3%, 3% and 2% for radish respectively for random method (averaged for 10 iterations), DUPLEX and CADEX algorithms. This was similar to the misclassification rate of 6% and 2% for cabbage and radish obtained using all 600 seeds in the calibration set. Thus, the number of seeds in the calibration set can be reduced by up to 67% without significant loss of classification accuracy, which will effectively enhance the cost-effectiveness of NIR spectral analysis. Wavelength regions important for the discrimination between viable and non-viable seeds were identified using interval ECVA (iECVA) models, ECVA weight plots and the mean difference spectrum for viable and no- viable seeds.",
keywords = "seeds, NIR, classification, ECVA, iECVA, PCA, DUPLEX, CADEX, misclassification rate ",
author = "Nisha Shetty and Tai-Gi Min and Ren{\'e} Gislum and Olesen, {Merete Halkj{\ae}r} and Birte Boelt",
year = "2011",
doi = "10.1255/jnirs.966",
language = "English",
volume = "19",
pages = "451--461",
journal = "Journal of Near Infrared Spectroscopy",
issn = "0967-0335",
publisher = "N I R Publications",
number = "6",

}

RIS

TY - JOUR

T1 - Optimal sample size for predicting viability of cabbage and radish seeds based on near infrared spectra of single seeds

AU - Shetty, Nisha

AU - Min, Tai-Gi

AU - Gislum, René

AU - Olesen, Merete Halkjær

AU - Boelt, Birte

PY - 2011

Y1 - 2011

N2 - The effects of the number of seeds in a training sample set on the ability to predict the viability of cabbage or radish seeds are presented and discussed. The supervised classification method extended canonical variates analysis (ECVA) was used to develop a classification model. Calibration sub-sets of different sizes were chosen randomly with several iterations and using the spectral-based sample selection algorithms DUPLEX and CADEX. An independent test set was used to validate the developed classification models. The results showed that 200 seeds were optimal in a calibration set for both cabbage and radish data. The misclassification rates at optimal sample size were 8%, 6% and 7% for cabbage and 3%, 3% and 2% for radish respectively for random method (averaged for 10 iterations), DUPLEX and CADEX algorithms. This was similar to the misclassification rate of 6% and 2% for cabbage and radish obtained using all 600 seeds in the calibration set. Thus, the number of seeds in the calibration set can be reduced by up to 67% without significant loss of classification accuracy, which will effectively enhance the cost-effectiveness of NIR spectral analysis. Wavelength regions important for the discrimination between viable and non-viable seeds were identified using interval ECVA (iECVA) models, ECVA weight plots and the mean difference spectrum for viable and no- viable seeds.

AB - The effects of the number of seeds in a training sample set on the ability to predict the viability of cabbage or radish seeds are presented and discussed. The supervised classification method extended canonical variates analysis (ECVA) was used to develop a classification model. Calibration sub-sets of different sizes were chosen randomly with several iterations and using the spectral-based sample selection algorithms DUPLEX and CADEX. An independent test set was used to validate the developed classification models. The results showed that 200 seeds were optimal in a calibration set for both cabbage and radish data. The misclassification rates at optimal sample size were 8%, 6% and 7% for cabbage and 3%, 3% and 2% for radish respectively for random method (averaged for 10 iterations), DUPLEX and CADEX algorithms. This was similar to the misclassification rate of 6% and 2% for cabbage and radish obtained using all 600 seeds in the calibration set. Thus, the number of seeds in the calibration set can be reduced by up to 67% without significant loss of classification accuracy, which will effectively enhance the cost-effectiveness of NIR spectral analysis. Wavelength regions important for the discrimination between viable and non-viable seeds were identified using interval ECVA (iECVA) models, ECVA weight plots and the mean difference spectrum for viable and no- viable seeds.

KW - seeds

KW - NIR

KW - classification

KW - ECVA

KW - iECVA

KW - PCA

KW - DUPLEX

KW - CADEX

KW - misclassification rate

U2 - 10.1255/jnirs.966

DO - 10.1255/jnirs.966

M3 - Journal article

VL - 19

SP - 451

EP - 461

JO - Journal of Near Infrared Spectroscopy

JF - Journal of Near Infrared Spectroscopy

SN - 0967-0335

IS - 6

ER -