- You are here:
-
-
All publications
- Near-infrared hyperspectral imaging for protein and hardness predictions of bulk samples of western canadian wheat from different locations and crop years using multivariate regression models
All publications from your selection
Near-infrared hyperspectral imaging for protein and hardness predictions of bulk samples of western canadian wheat from different locations and crop years using multivariate regression models
Authors: S. Mahesh, D. Jayas, J. Paliwal and N. White
Identifier: CSBE11119
Download file: https://library.csbe-scgab.ca/docs/meetings/2011/CSBE11119.pdf
Published in: CSBE-SCGAB Technical Conferences » AGM Winnipeg 2011
Download RAW file:
DOI:
Description: The objective of this study was to develop predictive multivariate regression models for estimating protein contents and hardness of bulk samples of wheat that were separated by locations and crop years. In this study, Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR) wheat samples were used. Wheat samples of different classes were obtained from nearby farms of different major wheat growing locations of Manitoba, Alberta, and Saskatchewan provinces and from 2007, 2008, 2009 crop years. The samples were conditioned to moisture levels of 13, 16, and 19% (wet basis) and pooled together for model development purposes. Image and spectral information of bulk samples in the spectral region between 960 and 1700 nm were acquired using the NIR hyperspectral imaging system at every 10 nm intervals. Reference protein contents and hardness values were determined using the Dumatherm method and single kernel characterization system (SKCS), respectively. In the protein content prediction study, using the full data set, the ten-component partial least squares regression (PLSR), which had 1.76%, 1.33%, and 0.68 for the estimated mean square error of prediction (MSEP), standard error of cross validation (SECV), and correlation coefficient (r), respectively, gave better results than the ten-component PCR model. In the hardness prediction part, the estimated MSEP, SECV, and r values were 147.7, 12.15, and 0.82, respectively, for the ten-component PLSR model. The NIR hyperspectral imaging can be used to predict quality parameters of wheat based on locations and crop years.
Keywords: wheat classes, protein, hardness, hyperspectral
Résumé:
Mots-clés:
Citation: S. Mahesh, D. Jayas, J. Paliwal and N. White. 2013. Near-infrared hyperspectral imaging for protein and hardness predictions of bulk samples of western canadian wheat from different locations and crop years using multivariate regression models . CSBE/SCGAB 2011 Annual Conference, Inn at the Forks, Winnipeg, Mb. 10-13 July 2011
Volume:
Issue:
Pages -
Contributor:
Date: juil-11
Technical field: Bioprocess Systems Engineering
Conference name: CSBE/SCGAB 2011 Annual Conference, Inn at the Forks, Winnipeg, Mb. 10-13 July 2011
Session name:
Other information:
Type: Text.Article
Format: PDF
Publication type: Conference Proceeding
Source:
Relation:
Coverage: Canada
Language 1: en
Language 2:
Rights: Canadian Society for Bioengineering
Notes:
Other files:
DOI:
Description: The objective of this study was to develop predictive multivariate regression models for estimating protein contents and hardness of bulk samples of wheat that were separated by locations and crop years. In this study, Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR) wheat samples were used. Wheat samples of different classes were obtained from nearby farms of different major wheat growing locations of Manitoba, Alberta, and Saskatchewan provinces and from 2007, 2008, 2009 crop years. The samples were conditioned to moisture levels of 13, 16, and 19% (wet basis) and pooled together for model development purposes. Image and spectral information of bulk samples in the spectral region between 960 and 1700 nm were acquired using the NIR hyperspectral imaging system at every 10 nm intervals. Reference protein contents and hardness values were determined using the Dumatherm method and single kernel characterization system (SKCS), respectively. In the protein content prediction study, using the full data set, the ten-component partial least squares regression (PLSR), which had 1.76%, 1.33%, and 0.68 for the estimated mean square error of prediction (MSEP), standard error of cross validation (SECV), and correlation coefficient (r), respectively, gave better results than the ten-component PCR model. In the hardness prediction part, the estimated MSEP, SECV, and r values were 147.7, 12.15, and 0.82, respectively, for the ten-component PLSR model. The NIR hyperspectral imaging can be used to predict quality parameters of wheat based on locations and crop years.
Keywords: wheat classes, protein, hardness, hyperspectral
Résumé:
Mots-clés:
Citation: S. Mahesh, D. Jayas, J. Paliwal and N. White. 2013. Near-infrared hyperspectral imaging for protein and hardness predictions of bulk samples of western canadian wheat from different locations and crop years using multivariate regression models . CSBE/SCGAB 2011 Annual Conference, Inn at the Forks, Winnipeg, Mb. 10-13 July 2011
Volume:
Issue:
Pages -
Contributor:
Date: juil-11
Technical field: Bioprocess Systems Engineering
Conference name: CSBE/SCGAB 2011 Annual Conference, Inn at the Forks, Winnipeg, Mb. 10-13 July 2011
Session name:
Other information:
Type: Text.Article
Format: PDF
Publication type: Conference Proceeding
Source:
Relation:
Coverage: Canada
Language 1: en
Language 2:
Rights: Canadian Society for Bioengineering
Notes:
Other files:
©
2024
CSBE-SCGAB