The Experts below are selected from a list of 294 Experts worldwide ranked by ideXlab platform
Noel D.g. White - One of the best experts on this subject based on the ideXlab platform.
-
Comparing Two Statistical Discriminant Models with a Back-Propagation Neural Network Model for Pairwise Classification of Location and Crop Year Specific Wheat Classes at Three Selected Moisture Contents Using NIR Hyperspectral Images
Transactions of the ASABE, 2014Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, Noel D.g. WhiteAbstract:Abstract. Knowledge of Wheat Classes and seed moisture contents not only determines the end use of Wheat flour but also helps in developing effective storage systems for Wheat. Samples of four Classes of Wheat, including Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR), were obtained from at least five different locations for each class in Manitoba, Saskatchewan, and Alberta for the 2007, 2008, and 2009 crop years and conditioned to moisture contents of 13%, 16%, and 19%. Near-infrared (NIR) hyperspectral images were acquired from bulk samples in the 960-1700 nm wavelength region at 10 nm intervals. The first and second principal component score images were compared for the segmented images of all Wheat Classes. Pairwise Wheat class identification was done using a non-parametric statistical model and a four-layer back-propagation neural network (BPNN) model. The NIR wavelengths of 1260 to 1380 nm had the highest factor loadings for the first principal component using principal component analysis (PCA). The four-layer BPNN model was used for two-class identification of Wheat Classes. Overall average pairwise classification accuracies of 83.7% were obtained for discriminating Wheat samples based on their moisture contents. Average classification accuracies of 83.2%, 75.4%, and 73.1%, were obtained for identifying Wheat Classes for samples with 13%, 16%, and 19% moisture content (m.c.), respectively. In this study, discriminant models yielded better classification accuracies than BPNN models. Overall average classification accuracies of Wheat Classes using statistical models were 80.6% for the linear discriminant analysis (LDA) and 76.3% for the quadratic discriminant analysis (QDA). This work showed that NIR hyperspectral imaging can be used as a potential nondestructive tool for classifying moisture-specific Wheat Classes.
-
Neural network prediction of Wheat Classes and moisture contents using near-infrared (NIR) hyperspectral images of bulk samples from different growing locations and crop years
2011 Louisville Kentucky August 7 - August 10 2011, 2011Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, Noel D.g. WhiteAbstract:Knowledge of Wheat Classes and moisture contents not only determines the end use of Wheat flour but also helps in developing effective storage systems for Wheat. Samples of four Classes of Wheat (Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR)) were obtained from five-six different locations in Manitoba, Saskatchewan, and Alberta for 2007, 2008, and 2009 crop years and conditioned to moisture contents of 13, 16, and 19%. Near-infrared hyperspectral images were acquired from bulk samples in the 960-1700 nm wavelength region at 10 nm intervals. Pair-wise average classification accuracies of 83.7 and 73.1-83.2% were obtained using a four-layer standard back propagation neural network (BPNN) model for identifying moisture contents and for discriminating Wheat Classes at each moisture level, respectively. Key wavelengths were identified based on their highest contributions towards classification. This work showed that NIR hyperspectral imaging can be used as a potential nondestructive tool for classifying moisture-specific Wheat Classes.
-
Characterization of the Influence of Moisture Content on the Morphological Features of Single Wheat Kernels Using Machine Vision
Applied Engineering in Agriculture, 2011Co-Authors: Ganesan Ramalingam, Digvir S. Jayas, Suresh Neethirajan, Noel D.g. WhiteAbstract:The objective of this study was to quantify changes in morphological features of kernels of western Canadian Wheat Classes caused by moisture increase using a machine vision system. One hundred single Wheat kernels for each of eight western Canadian Wheat Classes were successively conditioned from 12% to 20% (wet basis) moisture contents using potassium hydroxide (KOH) concentrations which regulated relative humidity. A digital camera of 7.4× 7.4-µm pixel resolution with an inter-line transfer charge-coupled device (CCD) image sensor was used to acquire images of single kernels. A machine vision algorithm developed at the Canadian Wheat Board Centre for Grain Storage Research, University of Manitoba, was implemented to extract seven morphological features (area, perimeter, major axis length, minor axis length, maximum radius, minimum radius, and mean radius) from the Wheat kernel images. All the seven features of Canada Western Red Spring, Canada Western Amber Durum, Canada Prairie Spring White, Canada Prairie Spring Red, Canada Western Extra Strong, Canada Western Red Winter, Canada Western Hard White Spring, and Canada Western Soft White Spring Wheat kernels were significantly (a = 0.05) different as the moisture content increased from 12% to 20%. All seven features showed a linearly increasing trend with an increase in moisture content.
-
Near-Infrared (NIR) Hyperspectral Imaging – An Emerging Analytical Tool for Classification of Western Canadian Wheat Classes from Different Locations and Crop Years
ASABE CSBE North Central Intersectional Meeting, 2010Co-Authors: Mahesh Sivakumar, Digvir S. Jayas, Jitendra Paliwal, Noel D.g. WhiteAbstract:A platform technology is needed for grain handling facilities in Canada to differentiate Wheat Classes. Near-infrared (NIR) hyperspectral imaging has recently emerged as a powerful analytical tool for conducting non-destructive quality analyses of agricultural and food samples. This study introduced a new analytical method using a NIR hyperspectral imaging system (960-1700 nm) to identify four western Canadian Wheat Classes at a uniform moisture level of 13%. Wheat samples used in this study were harvested during 2007, 2008, and 2009 crop years and collected from various growing locations in the prairie provinces (Manitoba, Saskatchewan, and Alberta) of Canada. Bulk samples of Wheat were scanned in the 960-1700 nm wavelength region at 10 nm intervals using an Indium Gallium Arsenide (InGaAs) NIR camera. The NIR reflectance intensities of scanned images were calculated and spectral data sets were created. Principal components analysis (PCA) was used to generate scores images and loadings plots. The versatility of the NIR hyperspectral imaging system was demonstrated using NIR reflectance and PCA scores images of samples of different Wheat Classes. The NIR wavelengths in the region of 1260-1380 nm had the highest factor loadings out of 75 wavelengths in the range of 960-1700 nm based on the first principal component. In statistical classification, the linear and quadratic discriminant classifiers had average classification accuracies of 95.4 and 92.3%, respectively, for identifying Wheat Classes that included sample variations due to growing locations and crop years. Also, wavelengths in the NIR regions of 1260-1380 and 1650-1700 nm were identified as important in classifying Wheat Classes, identifying growing locations, and determining crop years. This work showed that hyperspectral imaging technique can be used for rapidly identifying Wheat Classes.
-
Identification of Classes and Moisture Contents for Location-specific and Crop year-specific Wheat Samples Using the Near-Infrared (NIR) Hyperspectral Imaging
ASABE CSBE North Central Intersectional Meeting, 2010Co-Authors: Mahesh Sivakumar, Digvir S. Jayas, Jitendra Paliwal, Noel D.g. WhiteAbstract:Accurate segregation, proper drying, and safe storage can be possible when Wheat Classes and their moisture levels are properly identified. The potential of using a near infrared (NIR) hyperspectral imaging system to identify four western Canadian Wheat Classes each at 13, 16, and 19% moisture contents was investigated. Wheat samples harvested during 2007, 2008, and 2009 crop years were collected from various locations in Manitoba, Saskatchewan, and Alberta. An Indium Gallium Arsenide (InGaAs) NIR camera was used to scan bulk samples of Wheat in the 960-1700 nm wavelength region at 10 nm intervals. Calculated relative reflectance intensities of the scanned images were used for forming spectral data set. Scores images and loadings plots were generated using the principal component analysis (PCA). In both 16 and 19% moisture content (m.c.) Wheat, the highest factor loadings were in the region of 1260-1390 nm out of 75 NIR wavelengths in the 960-1700 nm range. In overall identification of four Wheat Classes independent growing locations, crop years, and moisture levels, average classification accuracies were of 80.6 and 76.3% for the linear discriminant analysis (LDA) and the quadratic discriminant analysis (QDA), respectively. For both 16 and 19% m.c. Wheat, wavelengths in the NIR regions of 1000-1200 and 1260-1390 nm were important in identifying Wheat Classes independent of growing locations and crop years. This study establishes that NIR hyperspectral imaging study can be used as a comprehensive tool for identifying Wheat Classes and moisture levels.
Jitendra Paliwal - One of the best experts on this subject based on the ideXlab platform.
-
comparison of partial least squares regression plsr and principal components regression pcr methods for protein and hardness predictions using the near infrared nir hyperspectral images of bulk samples of canadian Wheat
Food and Bioprocess Technology, 2015Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, N D G WhiteAbstract:The objective of this study was to compare the predictions of the protein contents and hardness values by partial least squares regression (PLSR) and principal components regression (PCR) models for bulk samples of Canadian Wheat, which were obtained from different locations and crop years. Wheat samples of Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR) Classes were obtained from nearby agricultural farms in the main Wheat growing locations in the Provinces of Alberta, Saskatchewan, and Manitoba from 2007, 2008, and 2009 crop years. Wheat samples were conditioned to moisture levels of 13, 16, and 19 % (wet basis) and pooled together for developing the regression models. A database of the near-infrared (NIR) hyperspectral image cubes of bulk samples of Wheat Classes was created in the wavelength region of 960–1,700 nm with 10 nm intervals. Reference protein contents and hardness values were determined using the Dumatherm method and single kernel characterization system (SKCS), respectively. A tenfold cross-validation was used for the ten-factor partial least squares regression (PLSR) and principal components regression (PCR) models for prediction purposes. Prediction performances of regression models were assessed by calculating the estimated mean square errors of prediction (MSEP), standard error of cross-validation (SECV), and correlation coefficient (r). Using the full data set in the protein prediction study, the ten-component PLSR model gave 1.76, 1.33, and 0.68 for the estimated MSEP, SECV, and r, respectively, which were better than the results for the ten-component PCR model (2.02, 1.42, and 0.62, respectively). For the hardness prediction, the estimated MSEP, SECV, and r values were 147.7, 12.15, and 0.82, respectively, for the ten-component PLSR model using the full data set. The PLSR models prediction performances outperformed the PCR models for predicting protein contents and hardness of Wheat.
-
Comparing Two Statistical Discriminant Models with a Back-Propagation Neural Network Model for Pairwise Classification of Location and Crop Year Specific Wheat Classes at Three Selected Moisture Contents Using NIR Hyperspectral Images
Transactions of the ASABE, 2014Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, Noel D.g. WhiteAbstract:Abstract. Knowledge of Wheat Classes and seed moisture contents not only determines the end use of Wheat flour but also helps in developing effective storage systems for Wheat. Samples of four Classes of Wheat, including Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR), were obtained from at least five different locations for each class in Manitoba, Saskatchewan, and Alberta for the 2007, 2008, and 2009 crop years and conditioned to moisture contents of 13%, 16%, and 19%. Near-infrared (NIR) hyperspectral images were acquired from bulk samples in the 960-1700 nm wavelength region at 10 nm intervals. The first and second principal component score images were compared for the segmented images of all Wheat Classes. Pairwise Wheat class identification was done using a non-parametric statistical model and a four-layer back-propagation neural network (BPNN) model. The NIR wavelengths of 1260 to 1380 nm had the highest factor loadings for the first principal component using principal component analysis (PCA). The four-layer BPNN model was used for two-class identification of Wheat Classes. Overall average pairwise classification accuracies of 83.7% were obtained for discriminating Wheat samples based on their moisture contents. Average classification accuracies of 83.2%, 75.4%, and 73.1%, were obtained for identifying Wheat Classes for samples with 13%, 16%, and 19% moisture content (m.c.), respectively. In this study, discriminant models yielded better classification accuracies than BPNN models. Overall average classification accuracies of Wheat Classes using statistical models were 80.6% for the linear discriminant analysis (LDA) and 76.3% for the quadratic discriminant analysis (QDA). This work showed that NIR hyperspectral imaging can be used as a potential nondestructive tool for classifying moisture-specific Wheat Classes.
-
Identification of Wheat Classes at different moisture levels using near-infrared hyperspectral images of bulk samples
Sensing and Instrumentation for Food Quality and Safety, 2011Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, N D G WhiteAbstract:Wheat Classes at different moisture levels need to be identified to accurately segregate, properly dry, and safely store before processing. This paper introduces a new method using a near infrared (NIR) hyperspectral imaging system (960–1,700 nm) to identify five western Canadian Wheat Classes (Canada Western Red Spring (CWRS), Canada Western Extra Strong (CWES), Canada Western Red Winter (CWRW), Canada Western Soft White Spring (CWSWS), and Canada Western Hard White Spring (CWHWS)) and moisture levels, independent of each other. The objectives of this research also included identification of each Wheat class at specific moisture levels of 12, 14, 16, 18, and 20%. Bulk samples of Wheat were scanned in the 960–1,700 nm wavelength region at 10 nm intervals using an Indium Gallium Arsenide (InGaAs) NIR camera. Spectral feature data sets were developed by calculating relative reflectance intensities of the scanned images. Principal components analysis was used to generate scores images and loadings plots. The NIR wavelengths in the region of 1,260–1,360 nm were important based on the loadings plot of first principal component. In statistical classification, the linear and quadratic discriminant analyses were used to classify Wheat Classes giving accuracies of 61–97 and 82–99%, respectively, independent of moisture contents. It was also found that the linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) could classify moisture contents with classification accuracies of 89–91 and 91–99%, respectively, independent of Wheat Classes. Once Wheat Classes were identified, classification accuracies of 90–100 and 72–99% were observed using LDA and QDA, respectively, when identifying specific moisture levels. Spectral features at key wavelengths of 1,060, 1,090, 1,340, and 1,450 nm were ranked at top in classifying Wheat Classes with different moisture contents. This work shows that hyperspectral imaging techniques can be used for rapidly identifying the Wheat Classes even at varying moisture levels.
-
Neural network prediction of Wheat Classes and moisture contents using near-infrared (NIR) hyperspectral images of bulk samples from different growing locations and crop years
2011 Louisville Kentucky August 7 - August 10 2011, 2011Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, Noel D.g. WhiteAbstract:Knowledge of Wheat Classes and moisture contents not only determines the end use of Wheat flour but also helps in developing effective storage systems for Wheat. Samples of four Classes of Wheat (Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR)) were obtained from five-six different locations in Manitoba, Saskatchewan, and Alberta for 2007, 2008, and 2009 crop years and conditioned to moisture contents of 13, 16, and 19%. Near-infrared hyperspectral images were acquired from bulk samples in the 960-1700 nm wavelength region at 10 nm intervals. Pair-wise average classification accuracies of 83.7 and 73.1-83.2% were obtained using a four-layer standard back propagation neural network (BPNN) model for identifying moisture contents and for discriminating Wheat Classes at each moisture level, respectively. Key wavelengths were identified based on their highest contributions towards classification. This work showed that NIR hyperspectral imaging can be used as a potential nondestructive tool for classifying moisture-specific Wheat Classes.
-
Near-Infrared (NIR) Hyperspectral Imaging – An Emerging Analytical Tool for Classification of Western Canadian Wheat Classes from Different Locations and Crop Years
ASABE CSBE North Central Intersectional Meeting, 2010Co-Authors: Mahesh Sivakumar, Digvir S. Jayas, Jitendra Paliwal, Noel D.g. WhiteAbstract:A platform technology is needed for grain handling facilities in Canada to differentiate Wheat Classes. Near-infrared (NIR) hyperspectral imaging has recently emerged as a powerful analytical tool for conducting non-destructive quality analyses of agricultural and food samples. This study introduced a new analytical method using a NIR hyperspectral imaging system (960-1700 nm) to identify four western Canadian Wheat Classes at a uniform moisture level of 13%. Wheat samples used in this study were harvested during 2007, 2008, and 2009 crop years and collected from various growing locations in the prairie provinces (Manitoba, Saskatchewan, and Alberta) of Canada. Bulk samples of Wheat were scanned in the 960-1700 nm wavelength region at 10 nm intervals using an Indium Gallium Arsenide (InGaAs) NIR camera. The NIR reflectance intensities of scanned images were calculated and spectral data sets were created. Principal components analysis (PCA) was used to generate scores images and loadings plots. The versatility of the NIR hyperspectral imaging system was demonstrated using NIR reflectance and PCA scores images of samples of different Wheat Classes. The NIR wavelengths in the region of 1260-1380 nm had the highest factor loadings out of 75 wavelengths in the range of 960-1700 nm based on the first principal component. In statistical classification, the linear and quadratic discriminant classifiers had average classification accuracies of 95.4 and 92.3%, respectively, for identifying Wheat Classes that included sample variations due to growing locations and crop years. Also, wavelengths in the NIR regions of 1260-1380 and 1650-1700 nm were identified as important in classifying Wheat Classes, identifying growing locations, and determining crop years. This work showed that hyperspectral imaging technique can be used for rapidly identifying Wheat Classes.
Digvir S. Jayas - One of the best experts on this subject based on the ideXlab platform.
-
comparison of partial least squares regression plsr and principal components regression pcr methods for protein and hardness predictions using the near infrared nir hyperspectral images of bulk samples of canadian Wheat
Food and Bioprocess Technology, 2015Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, N D G WhiteAbstract:The objective of this study was to compare the predictions of the protein contents and hardness values by partial least squares regression (PLSR) and principal components regression (PCR) models for bulk samples of Canadian Wheat, which were obtained from different locations and crop years. Wheat samples of Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR) Classes were obtained from nearby agricultural farms in the main Wheat growing locations in the Provinces of Alberta, Saskatchewan, and Manitoba from 2007, 2008, and 2009 crop years. Wheat samples were conditioned to moisture levels of 13, 16, and 19 % (wet basis) and pooled together for developing the regression models. A database of the near-infrared (NIR) hyperspectral image cubes of bulk samples of Wheat Classes was created in the wavelength region of 960–1,700 nm with 10 nm intervals. Reference protein contents and hardness values were determined using the Dumatherm method and single kernel characterization system (SKCS), respectively. A tenfold cross-validation was used for the ten-factor partial least squares regression (PLSR) and principal components regression (PCR) models for prediction purposes. Prediction performances of regression models were assessed by calculating the estimated mean square errors of prediction (MSEP), standard error of cross-validation (SECV), and correlation coefficient (r). Using the full data set in the protein prediction study, the ten-component PLSR model gave 1.76, 1.33, and 0.68 for the estimated MSEP, SECV, and r, respectively, which were better than the results for the ten-component PCR model (2.02, 1.42, and 0.62, respectively). For the hardness prediction, the estimated MSEP, SECV, and r values were 147.7, 12.15, and 0.82, respectively, for the ten-component PLSR model using the full data set. The PLSR models prediction performances outperformed the PCR models for predicting protein contents and hardness of Wheat.
-
Comparing Two Statistical Discriminant Models with a Back-Propagation Neural Network Model for Pairwise Classification of Location and Crop Year Specific Wheat Classes at Three Selected Moisture Contents Using NIR Hyperspectral Images
Transactions of the ASABE, 2014Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, Noel D.g. WhiteAbstract:Abstract. Knowledge of Wheat Classes and seed moisture contents not only determines the end use of Wheat flour but also helps in developing effective storage systems for Wheat. Samples of four Classes of Wheat, including Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR), were obtained from at least five different locations for each class in Manitoba, Saskatchewan, and Alberta for the 2007, 2008, and 2009 crop years and conditioned to moisture contents of 13%, 16%, and 19%. Near-infrared (NIR) hyperspectral images were acquired from bulk samples in the 960-1700 nm wavelength region at 10 nm intervals. The first and second principal component score images were compared for the segmented images of all Wheat Classes. Pairwise Wheat class identification was done using a non-parametric statistical model and a four-layer back-propagation neural network (BPNN) model. The NIR wavelengths of 1260 to 1380 nm had the highest factor loadings for the first principal component using principal component analysis (PCA). The four-layer BPNN model was used for two-class identification of Wheat Classes. Overall average pairwise classification accuracies of 83.7% were obtained for discriminating Wheat samples based on their moisture contents. Average classification accuracies of 83.2%, 75.4%, and 73.1%, were obtained for identifying Wheat Classes for samples with 13%, 16%, and 19% moisture content (m.c.), respectively. In this study, discriminant models yielded better classification accuracies than BPNN models. Overall average classification accuracies of Wheat Classes using statistical models were 80.6% for the linear discriminant analysis (LDA) and 76.3% for the quadratic discriminant analysis (QDA). This work showed that NIR hyperspectral imaging can be used as a potential nondestructive tool for classifying moisture-specific Wheat Classes.
-
Identification of Wheat Classes at different moisture levels using near-infrared hyperspectral images of bulk samples
Sensing and Instrumentation for Food Quality and Safety, 2011Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, N D G WhiteAbstract:Wheat Classes at different moisture levels need to be identified to accurately segregate, properly dry, and safely store before processing. This paper introduces a new method using a near infrared (NIR) hyperspectral imaging system (960–1,700 nm) to identify five western Canadian Wheat Classes (Canada Western Red Spring (CWRS), Canada Western Extra Strong (CWES), Canada Western Red Winter (CWRW), Canada Western Soft White Spring (CWSWS), and Canada Western Hard White Spring (CWHWS)) and moisture levels, independent of each other. The objectives of this research also included identification of each Wheat class at specific moisture levels of 12, 14, 16, 18, and 20%. Bulk samples of Wheat were scanned in the 960–1,700 nm wavelength region at 10 nm intervals using an Indium Gallium Arsenide (InGaAs) NIR camera. Spectral feature data sets were developed by calculating relative reflectance intensities of the scanned images. Principal components analysis was used to generate scores images and loadings plots. The NIR wavelengths in the region of 1,260–1,360 nm were important based on the loadings plot of first principal component. In statistical classification, the linear and quadratic discriminant analyses were used to classify Wheat Classes giving accuracies of 61–97 and 82–99%, respectively, independent of moisture contents. It was also found that the linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) could classify moisture contents with classification accuracies of 89–91 and 91–99%, respectively, independent of Wheat Classes. Once Wheat Classes were identified, classification accuracies of 90–100 and 72–99% were observed using LDA and QDA, respectively, when identifying specific moisture levels. Spectral features at key wavelengths of 1,060, 1,090, 1,340, and 1,450 nm were ranked at top in classifying Wheat Classes with different moisture contents. This work shows that hyperspectral imaging techniques can be used for rapidly identifying the Wheat Classes even at varying moisture levels.
-
Characterization of the Influence of Moisture Content on the Morphological Features of Single Wheat Kernels Using Machine Vision
Applied Engineering in Agriculture, 2011Co-Authors: Ganesan Ramalingam, Digvir S. Jayas, Suresh Neethirajan, Noel D.g. WhiteAbstract:The objective of this study was to quantify changes in morphological features of kernels of western Canadian Wheat Classes caused by moisture increase using a machine vision system. One hundred single Wheat kernels for each of eight western Canadian Wheat Classes were successively conditioned from 12% to 20% (wet basis) moisture contents using potassium hydroxide (KOH) concentrations which regulated relative humidity. A digital camera of 7.4× 7.4-µm pixel resolution with an inter-line transfer charge-coupled device (CCD) image sensor was used to acquire images of single kernels. A machine vision algorithm developed at the Canadian Wheat Board Centre for Grain Storage Research, University of Manitoba, was implemented to extract seven morphological features (area, perimeter, major axis length, minor axis length, maximum radius, minimum radius, and mean radius) from the Wheat kernel images. All the seven features of Canada Western Red Spring, Canada Western Amber Durum, Canada Prairie Spring White, Canada Prairie Spring Red, Canada Western Extra Strong, Canada Western Red Winter, Canada Western Hard White Spring, and Canada Western Soft White Spring Wheat kernels were significantly (a = 0.05) different as the moisture content increased from 12% to 20%. All seven features showed a linearly increasing trend with an increase in moisture content.
-
Neural network prediction of Wheat Classes and moisture contents using near-infrared (NIR) hyperspectral images of bulk samples from different growing locations and crop years
2011 Louisville Kentucky August 7 - August 10 2011, 2011Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, Noel D.g. WhiteAbstract:Knowledge of Wheat Classes and moisture contents not only determines the end use of Wheat flour but also helps in developing effective storage systems for Wheat. Samples of four Classes of Wheat (Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR)) were obtained from five-six different locations in Manitoba, Saskatchewan, and Alberta for 2007, 2008, and 2009 crop years and conditioned to moisture contents of 13, 16, and 19%. Near-infrared hyperspectral images were acquired from bulk samples in the 960-1700 nm wavelength region at 10 nm intervals. Pair-wise average classification accuracies of 83.7 and 73.1-83.2% were obtained using a four-layer standard back propagation neural network (BPNN) model for identifying moisture contents and for discriminating Wheat Classes at each moisture level, respectively. Key wavelengths were identified based on their highest contributions towards classification. This work showed that NIR hyperspectral imaging can be used as a potential nondestructive tool for classifying moisture-specific Wheat Classes.
Annamalai Manickavasagan - One of the best experts on this subject based on the ideXlab platform.
-
Comparison of illuminations to identify Wheat Classes using monochrome images
Computers and Electronics in Agriculture, 2008Co-Authors: Annamalai Manickavasagan, G. Sathya, Digvir S. JayasAbstract:Wheat class identification using machine vision is an objective method which can be used for online testing to automate handling, binning and shipping operations in grain industry. The efficiencies of a monochrome camera-based vision system with three different illuminations (incandescent light (IL), fluorescent ring light (FRL), fluorescent tube light (FTL)) were determined to identify eight western Canadian Wheat Classes at four moisture levels (11%, 14%, 17% and 20%). The monochrome images of the bulk Wheat samples were acquired at each moisture level (3 illuminationsx8 Classesx4 moisture contentsx100 replications=9600 images). A linear discriminant function was used for the classification of Wheat samples using 32 gray level textural features extracted from the monochrome images. The mean gray values of the Wheat Classes were in the ranges of 75-103, 73-115, and 107-143 for IL, FRL and FTL, respectively. The mean gray values of Wheat Classes were significantly different within each illumination and between different illuminations (@a=0.05). Mean gray value was the highest for FTL and the lowest for IL illumination. The moisture content of the Wheat samples had significant effect on the mean gray values. The overall classification accuracies were 90%, 81% and 96% for IL, FRL and FTL, respectively, when all the Wheat Classes were at the same moisture levels. It was 66%, 53% and 85% for IL, FRL and FTL, respectively, when the Wheat Classes were at different moisture levels. The classification accuracies of a 2-stage classification system for the Classes with different moisture levels were 68%, 56% and 90% for IL, FRL and FTL, respectively.
-
feasibility of near infrared hyperspectral imaging to differentiate canadian Wheat Classes
Biosystems Engineering, 2008Co-Authors: S Mahesh, Annamalai Manickavasagan, Digvir S. Jayas, Jitendra Paliwal, N D G WhiteAbstract:Differentiation of Wheat Classes is one of the important challenges to the Canadian grain industry. Even though some Wheat Classes may look similar, their chemical composition and consequently the end-product quality can vary significantly. Visual differentiation of Wheat Classes suffers from disadvantages such as inconsistency, low throughput, and labour intensiveness. A near-infrared (NIR) hyperspectral imaging system was used to develop classification models to differentiate Wheat Classes grown in western Canada. Wheat bulk samples were scanned in the wavelength region of 960–1700 nm at 10 nm intervals using an InGaAs NIR camera. Seventy-five relative reflectance intensities were extracted from the scanned images and used for the differentiation of Wheat Classes using a statistical classifier and an artificial neural network (ANN) classifier. Classification accuracies were 100% in classifying Canada Prairie Spring Red (CPSR), Canada Western Red Winter (CWRW), and Canada Western Soft White Spring (CWSWS) Wheat Classes and >94% for the other Wheat Classes (Canada Western Extra Strong (CWES), Canada Western Hard White Spring (CWHWS), Canada Western Red Spring (CWRS), Canada Prairie Spring White (CPSW) and Canada Western Amber Durum (CWAD)) using Linear Discriminant Analysis (LDA) with a leave-one-out cross-validation method. In Quadratic Discriminant Analysis (QDA) with a leave-one-out cross-validation method, the classification accuracies were >86% for all Wheat Classes. The overall classification accuracies of 60% training–30% testing–10% validation (referred to as 60-30-10) and 70% training–20% testing–10% validation (referred to as 70-20-10) ANN models were above 90% for independent validation sets using three-layer standard and Wardnet back-propagation neural network architectures.
-
Wheat Class Identification Using Thermal Imaging
Food and Bioprocess Technology, 2008Co-Authors: Annamalai Manickavasagan, Digvir S. Jayas, Noel D.g. White, Jitendra PaliwalAbstract:Wheat Classes and varieties are determined by trained professionals in the laboratory. Several approaches have been made using machine vision technology for nondestructive and online identification of Wheat Classes, but the performance has been poor and inconsistent. An infrared thermal imaging system was developed to identify eight western Canadian Wheat Classes. Samples of 20 g each of Wheat at 14% moisture content (wet basis) spread in a 100 × 100 mm monolayer were heated by a plate heater (maintained at 90 °C) placed at a distance of 10 mm from the grain layer. The surface temperatures of the top surface of the grain bulk were imaged before heating, after heating for 180 s, and after cooling for 30 s using an infrared thermal camera (n = 100). Temperature rise (after heating) and drop (after cooling) were significantly different for Wheat Classes (α = 0.05). The temperature rise ranged from 14.94 (Canada Western Red Spring) to 17.80 °C (Canada Prairie Spring Red), and the drop ranged from 3.67 (Canada Western Extra Strong) to 4.42 °C (Canada Prairie Spring Red) after heating for 180 s and cooling for 30 s, respectively. The rate of heating and cooling was negatively correlated with protein content of Wheat (r = −0.63 for heating, r = −0.65 for cooling) and true density (r = −0.67 for heating, r = −0.71 for cooling), and positively correlated with grain hardness (r = +0.41 for heating, r = +0.53 for cooling). Overall classification accuracies of an eight-class model, red-class model (four Classes), white-class model (four Classes), and pairwise (two-class model) comparisons using a quadratic discriminant method were 76%, 87%, 79%, and 95%, and 64%, 87%, 77%, and 91% using bootstrap and leave-one-out validation methods, respectively. There were several misclassifications in the four and eight-class models. Thermal imaging approach may have potential to develop classification methods for two Classes, which are similar and difficult to distinguish by visual inspection; however, the effect of growing season, defects, and kernel size must be considered while developing such methods. The temperature rise after heating and drop after cooling were tested for Canada Western Red Spring Wheat at three moisture levels (11%, 14%, and 17% wet basis; n = 20). There were no significant differences (α = 0.05) in the mean temperature rise and temperature drop between 11%, 14%, and 17% moisture samples.
-
Wheat class identification using monochrome images
Journal of Cereal Science, 2008Co-Authors: Annamalai Manickavasagan, G. Sathya, Digvir S. Jayas, Noel D.g. WhiteAbstract:Abstract Wheat class identification by bulk sample analysis using a machine vision method would be helpful for automation of grain handling, binning and shipping operations in grain elevators. A machine vision system with a monochrome camera was used to identify eight western Canadian Wheat Classes at four moisture levels (11%, 14%, 17% and 20% wet basis) by bulk sample analysis (n=100 images for each group of samples). Grayscale images (1024×768 pixels) of the grain bulk were captured by the monochrome camera, and stored on a data acquisition system. Algorithms were developed to extract 32 textural features automatically from the grayscale images. The mean gray values of the western Canadian Wheat Classes ranged between 106 and 143, and it was the highest for Canada Prairie Spring Red and the lowest for Canada Western Extra Strong and Canada Western Red Winter. The mean gray values of the Wheat samples were significantly higher at 17% moisture content and lower at 11% moisture content among the tested moisture levels (α=0.05). The overall classification accuracies of a quadratic discriminant function were 93.8%, 92.5%, 92.0% and 94.4% when the Wheat Classes were at 11%, 14%, 17% and 20% moisture contents, respectively. Similarly, the accuracies of a linear discriminant function were 96.1%, 95.0%, 95.4% and 96.3% at 11%, 14%, 17% and 20% moisture contents, respectively. When the Wheat Classes were identified irrespective of moisture levels (images of the four moisture level grains in each class were mixed together), the accuracy was 89.8% and 85.4% for quadratic and linear discriminant functions, respectively. A monochrome image analysis system has the potential to use for online identification of Classes in Wheat handling facilities. However, further research is required to determine the performance of the developed method for impurities in bulk grain such as foreign material and dockage.
-
Feasibility of near-infrared hyperspectral imaging to differentiate Canadian Wheat Classes
Biosystems Engineering, 2008Co-Authors: S Mahesh, Annamalai Manickavasagan, Digvir S. Jayas, Jitendra Paliwal, Noel D.g. WhiteAbstract:Differentiation of Wheat Classes is one of the important challenges to the Canadian grain industry. Even though some Wheat Classes may look similar, their chemical composition and consequently the end-product quality can vary significantly. Visual differentiation of Wheat Classes suffers from disadvantages such as inconsistency, low throughput, and labour intensiveness. A near-infrared (NIR) hyperspectral imaging system was used to develop classification models to differentiate Wheat Classes grown in western Canada. Wheat bulk samples were scanned in the wavelength region of 960–1700 nm at 10 nm intervals using an InGaAs NIR camera. Seventy-five relative reflectance intensities were extracted from the scanned images and used for the differentiation of Wheat Classes using a statistical classifier and an artificial neural network (ANN) classifier. Classification accuracies were 100% in classifying Canada Prairie Spring Red (CPSR), Canada Western Red Winter (CWRW), and Canada Western Soft White Spring (CWSWS) Wheat Classes and >94% for the other Wheat Classes (Canada Western Extra Strong (CWES), Canada Western Hard White Spring (CWHWS), Canada Western Red Spring (CWRS), Canada Prairie Spring White (CPSW) and Canada Western Amber Durum (CWAD)) using Linear Discriminant Analysis (LDA) with a leave-one-out cross-validation method. In Quadratic Discriminant Analysis (QDA) with a leave-one-out cross-validation method, the classification accuracies were >86% for all Wheat Classes. The overall classification accuracies of 60% training–30% testing–10% validation (referred to as 60-30-10) and 70% training–20% testing–10% validation (referred to as 70-20-10) ANN models were above 90% for independent validation sets using three-layer standard and Wardnet back-propagation neural network architectures.
N D G White - One of the best experts on this subject based on the ideXlab platform.
-
comparison of partial least squares regression plsr and principal components regression pcr methods for protein and hardness predictions using the near infrared nir hyperspectral images of bulk samples of canadian Wheat
Food and Bioprocess Technology, 2015Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, N D G WhiteAbstract:The objective of this study was to compare the predictions of the protein contents and hardness values by partial least squares regression (PLSR) and principal components regression (PCR) models for bulk samples of Canadian Wheat, which were obtained from different locations and crop years. Wheat samples of Canada Western Red Spring (CWRS), Canada Western Hard White Spring (CWHWS), Canada Western Soft White Spring (CWSWS), and Canada Prairie Spring Red (CPSR) Classes were obtained from nearby agricultural farms in the main Wheat growing locations in the Provinces of Alberta, Saskatchewan, and Manitoba from 2007, 2008, and 2009 crop years. Wheat samples were conditioned to moisture levels of 13, 16, and 19 % (wet basis) and pooled together for developing the regression models. A database of the near-infrared (NIR) hyperspectral image cubes of bulk samples of Wheat Classes was created in the wavelength region of 960–1,700 nm with 10 nm intervals. Reference protein contents and hardness values were determined using the Dumatherm method and single kernel characterization system (SKCS), respectively. A tenfold cross-validation was used for the ten-factor partial least squares regression (PLSR) and principal components regression (PCR) models for prediction purposes. Prediction performances of regression models were assessed by calculating the estimated mean square errors of prediction (MSEP), standard error of cross-validation (SECV), and correlation coefficient (r). Using the full data set in the protein prediction study, the ten-component PLSR model gave 1.76, 1.33, and 0.68 for the estimated MSEP, SECV, and r, respectively, which were better than the results for the ten-component PCR model (2.02, 1.42, and 0.62, respectively). For the hardness prediction, the estimated MSEP, SECV, and r values were 147.7, 12.15, and 0.82, respectively, for the ten-component PLSR model using the full data set. The PLSR models prediction performances outperformed the PCR models for predicting protein contents and hardness of Wheat.
-
Identification of Wheat Classes at different moisture levels using near-infrared hyperspectral images of bulk samples
Sensing and Instrumentation for Food Quality and Safety, 2011Co-Authors: S Mahesh, Digvir S. Jayas, Jitendra Paliwal, N D G WhiteAbstract:Wheat Classes at different moisture levels need to be identified to accurately segregate, properly dry, and safely store before processing. This paper introduces a new method using a near infrared (NIR) hyperspectral imaging system (960–1,700 nm) to identify five western Canadian Wheat Classes (Canada Western Red Spring (CWRS), Canada Western Extra Strong (CWES), Canada Western Red Winter (CWRW), Canada Western Soft White Spring (CWSWS), and Canada Western Hard White Spring (CWHWS)) and moisture levels, independent of each other. The objectives of this research also included identification of each Wheat class at specific moisture levels of 12, 14, 16, 18, and 20%. Bulk samples of Wheat were scanned in the 960–1,700 nm wavelength region at 10 nm intervals using an Indium Gallium Arsenide (InGaAs) NIR camera. Spectral feature data sets were developed by calculating relative reflectance intensities of the scanned images. Principal components analysis was used to generate scores images and loadings plots. The NIR wavelengths in the region of 1,260–1,360 nm were important based on the loadings plot of first principal component. In statistical classification, the linear and quadratic discriminant analyses were used to classify Wheat Classes giving accuracies of 61–97 and 82–99%, respectively, independent of moisture contents. It was also found that the linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) could classify moisture contents with classification accuracies of 89–91 and 91–99%, respectively, independent of Wheat Classes. Once Wheat Classes were identified, classification accuracies of 90–100 and 72–99% were observed using LDA and QDA, respectively, when identifying specific moisture levels. Spectral features at key wavelengths of 1,060, 1,090, 1,340, and 1,450 nm were ranked at top in classifying Wheat Classes with different moisture contents. This work shows that hyperspectral imaging techniques can be used for rapidly identifying the Wheat Classes even at varying moisture levels.
-
near infrared nir hyperspectral imaging an emerging analytical tool for classification of western canadian Wheat Classes from different locations and crop years
ASABE CSBE North Central Intersectional Meeting, 2010Co-Authors: Mahesh Sivakumar, Digvir S. Jayas, Jitendra Paliwal, N D G WhiteAbstract:A platform technology is needed for grain handling facilities in Canada to differentiate Wheat Classes. Near-infrared (NIR) hyperspectral imaging has recently emerged as a powerful analytical tool for conducting non-destructive quality analyses of agricultural and food samples. This study introduced a new analytical method using a NIR hyperspectral imaging system (960-1700 nm) to identify four western Canadian Wheat Classes at a uniform moisture level of 13%. Wheat samples used in this study were harvested during 2007, 2008, and 2009 crop years and collected from various growing locations in the prairie provinces (Manitoba, Saskatchewan, and Alberta) of Canada. Bulk samples of Wheat were scanned in the 960-1700 nm wavelength region at 10 nm intervals using an Indium Gallium Arsenide (InGaAs) NIR camera. The NIR reflectance intensities of scanned images were calculated and spectral data sets were created. Principal components analysis (PCA) was used to generate scores images and loadings plots. The versatility of the NIR hyperspectral imaging system was demonstrated using NIR reflectance and PCA scores images of samples of different Wheat Classes. The NIR wavelengths in the region of 1260-1380 nm had the highest factor loadings out of 75 wavelengths in the range of 960-1700 nm based on the first principal component. In statistical classification, the linear and quadratic discriminant classifiers had average classification accuracies of 95.4 and 92.3%, respectively, for identifying Wheat Classes that included sample variations due to growing locations and crop years. Also, wavelengths in the NIR regions of 1260-1380 and 1650-1700 nm were identified as important in classifying Wheat Classes, identifying growing locations, and determining crop years. This work showed that hyperspectral imaging technique can be used for rapidly identifying Wheat Classes.
-
feasibility of near infrared hyperspectral imaging to differentiate canadian Wheat Classes
Biosystems Engineering, 2008Co-Authors: S Mahesh, Annamalai Manickavasagan, Digvir S. Jayas, Jitendra Paliwal, N D G WhiteAbstract:Differentiation of Wheat Classes is one of the important challenges to the Canadian grain industry. Even though some Wheat Classes may look similar, their chemical composition and consequently the end-product quality can vary significantly. Visual differentiation of Wheat Classes suffers from disadvantages such as inconsistency, low throughput, and labour intensiveness. A near-infrared (NIR) hyperspectral imaging system was used to develop classification models to differentiate Wheat Classes grown in western Canada. Wheat bulk samples were scanned in the wavelength region of 960–1700 nm at 10 nm intervals using an InGaAs NIR camera. Seventy-five relative reflectance intensities were extracted from the scanned images and used for the differentiation of Wheat Classes using a statistical classifier and an artificial neural network (ANN) classifier. Classification accuracies were 100% in classifying Canada Prairie Spring Red (CPSR), Canada Western Red Winter (CWRW), and Canada Western Soft White Spring (CWSWS) Wheat Classes and >94% for the other Wheat Classes (Canada Western Extra Strong (CWES), Canada Western Hard White Spring (CWHWS), Canada Western Red Spring (CWRS), Canada Prairie Spring White (CPSW) and Canada Western Amber Durum (CWAD)) using Linear Discriminant Analysis (LDA) with a leave-one-out cross-validation method. In Quadratic Discriminant Analysis (QDA) with a leave-one-out cross-validation method, the classification accuracies were >86% for all Wheat Classes. The overall classification accuracies of 60% training–30% testing–10% validation (referred to as 60-30-10) and 70% training–20% testing–10% validation (referred to as 70-20-10) ANN models were above 90% for independent validation sets using three-layer standard and Wardnet back-propagation neural network architectures.
