Application of image analysis for the varietal classification of barley:: Morphological features

This paper presents an exploratory investigation of the application of morphological features in image analysis for varietal classification of Polish spring barley. The objective of this study was to determine the utility of morphological features for classifying individual kernels of five varieties of barley. Furthermore, this study was performed to find the best method to classify kernels of barley with the lowest error of classification. Image processing consisted of several steps: image acquisition, segmentation, external and internal image feature extraction, classification and interpretation. Each barley kernel was described using 74 morphological features. The selection was carried out using three methods based on: Fisher's coefficient, probability of error and average correlation coefficient and mutual information. Principle component analysis (PCA), linear discriminant analysis (LDA), and non-linear discriminant analysis (NDA) were used throughout this paper as the classification methods. The results confirmed that the method using morphological features may be successfully employed in image analysis for preliminary varietal identification of barley kernels. Furthermore, LDA was found to be the method which best separated different varieties of objects.     

Study of the microstructure of durum wheat endosperm using X-ray micro-computed tomography

Durum wheat grains are used for producing food, such as pasta or couscous. The grain mechanical properties which are linked to its internal micro-structure (i.e. endosperm porosity) are known to determine its ability to produce semolina during milling. The proportion of grains having porous endosperm in a batch appears therefore as a critical quality factor for the durum wheat value chain. Our objective was to investigate the ability of X-ray micro-tomography (μCT) method to describe the porous or vitreous counterpart structures in the endosperm of durum wheat grains. We selected two different durum wheat samples displaying vitreous or partially porous endosperms. The grains were analyzed using μCT at two pixel sizes (1 μm or 7 μm). The μCT data collected at 7 μm pixel size were used for qualitative classification of grains according to apparent distribution curve of the porosity parameters. Analysis of μCT images at 1 μm pixel size allowed us to propose pore size classification in the vitreous and porous parts of the endosperm in three durum wheat grain. Results are used to better describe the durum-wheat endosperm microstructure, but requires long scanning periods.     

基于SVM的县域冬小麦种植面积遥感提取

冬小麦种植面积的精确提取,对于农业部门进行冬小麦生长监测与产量估测有着重要的支撑作用。本研究在对Landsat-8卫星15 m×15 m空间分辨率遥感影像进行预处理的基础上,基于最佳波段指数(OIF),采用支持向量机(SVM)算法中四种核函数进行影像分类,并比较分类精度,选择精度最高的核函数作为SVM最优核函数对盐城市大丰区冬小麦种植面积进行提取,与最大似然法、最小距离法的结果进行对比。结果表明,四种核函数中,Linear核函数分类精度最高,达到98.56%。将Linear核函数作为SVM最优核函数对大丰区冬小麦种植面积进行提取,提取到的种植面积为71 834.4 hm~2,提取精度、分类精度和Kappa系数分别为91.25%、98.56%和0.98。基于SVM的冬小麦面积提取效果明显好于传统监督分类方法,说明使用支持向量机与影像光谱特征进行影像分类能够准确提取县域冬小麦种植面积。     

1H NMR-based metabolomics for discrimination of rice from different geographical origins of China

Food frauds related to the mislabeling and mixing of products of inferior quality with those of superior quality are a serious concern nowadays. NMR-based metabolomics has great potential in the authentication of foods for quality assurance and the tracing of fraudulent labeling. The present study was conducted to discriminate rice from geographically different provinces of China. The study reports the potential use of ¹H NMR spectroscopy coupled with PCA and a discriminant analysis method, LDA for metabolomic fingerprinting of Chinese rice. A total of 106 rice samples from nine different provinces of China were analyzed for ¹H NMR-based metabolomics. Both the whole variable analysis (heat map) and the Principal Component Analysis (PCA) showed a clear separation among the samples. Linear Discriminant Analysis (LDA) was conducted to extract the variables majorly responsible for this separation, such as sucrose, fructose, glucose, succinate, polyphenols, trigonelline and asparagine. The discrimination was explained on the basis of variations in latitude, temperature and rainfall in these provinces. The study highlights the application of ¹H NMR for geographical discrimination of rice and its usefulness for consumers while choosing their desired variety of rice.     

A two-camera machine vision in predicting alpha-amylase activity in wheat

Sprout damage in wheat is a serious problem worldwide because damaged wheat kernels contain alpha-amylase, an enzyme that causes poor baking quality of wheat. A two-camera machine vision (MV) with a neural network was implemented to quantify alpha-amylase activity in wheat using 16 visual properties of the kernels. Kernels were separated at image level using the marker-controlled segmentation algorithm before the properties (color, texture, and shape and size) of dorsal and ventral sides of kernels were extracted. Alpha-amylase activity in wheat was assessed analytically. The neural networks were trained, validated, and tested using the visual properties as the inputs and alpha-amylase activity as the output. The trained neural network predicted alpha-amylase activity with an accuracy of 6913 U/L (rmse) and R² value of 0.72 for the wheat samples with alpha-amylase activity ranging over 178 to 28935 (U/L). Differences between visual properties of wheat samples calculated from the top and the bottom images was less than 0.5%. Light stability in time and influence of temperature on the cameras' color stability were less than 2% of the mean values. The challenges associated with the system, and recommendations to improve the system accuracy and robustness, and to decrease the system cost are presented.     

Opaque2基因对糯玉米子粒品质的影响分析

玉米Opaque2(O2)基因突变显著影响子粒胚乳中蛋白体形成、淀粉含量、氨基酸组成等,其中,赖氨酸含量提高使营养品质大幅度提升。通过回交转育的方法构建17份糯玉米o2近等基因系,对子粒表型、种皮厚度、百粒重和粗淀粉含量等性状分析。结果表明,与对照相比,16个近等基因系子粒明显皱缩、胚乳变为粉质,6个近等基因系种皮厚度增加,11个近等基因系百粒重显著降低,12个近等基因系子粒粗淀粉含量显著降低。由此表明,o2基因的导入对糯玉米子粒表型、产量、淀粉含量等主要起负调控作用,影响程度与导入背景密切相关。     

Evaluation and analysis the chalkiness of connected rice kernels based on image processing technology and support vector machine

In order to determine the location and type of rice chalkiness accurately, image processing techniques were adopted to process acquired rice kernel images. Connected rice kernels were separated from each other using a convex point matching method. Chalkiness was extracted according to the differences in grayscale levels between chalky and normal regions in the rice kernel and chalky rice kernels were classified by a support vector machine (SVM). The results showed that 2–5 connected rice kernels could be separated accurately using this method and chalky areas could be extracted. The classification accuracy for indica rice and japonica rice reached 98.5% and 97.6%, respectively, by using SVM. Hence, the measurement results are accurate and reliable, and the presented work provides a theoretical and practical basis for the further application of computer vision technology to chalkiness detection.     

Effects of Maize Type and Fermentation Conditions on the Quality of Beninese Traditional Ogi,a Fermented Maize Slurry

Four Beninese cultivars (two local ecotypes and two new cultivars) were tested for their ability to produce ogi. The two local ecotypes, having friable and floury to medium vitreous kernels, gave a higher yield of ogi (10% more) than the improved cultivars that had coherent and vitreous kernels. They also gave slightly more viscous ogi porridges. Whatever the storage conditions (20 to 35°C, 3 to 15 d), a hetero-lactic fermentation generated two characteristic products: lactic acid and ethanol. However, the fermentation process was more intense and more rapid at 35°C giving a peak in acidity at the end of the first week, then a decrease during the second week. In parallel, ori porridge viscosity increased during the second week of fermentation.     

Quantification and distribution of kafirins in the kernels of sorghum cultivars varying in endosperm hardness

Kafirins extracted from the endosperm of seven sorghum [Sorghum bicolor (L.) Moench] cultivars were analysed by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) and enzyme-linked immunosorbent assay (ELISA). Kafirins extracted from the vitreous and floury endosperm portions within the kernel were also analysed by these techniques. The ELISA results indicated that the level of all the three kafirins was high in the hard endosperm kernels. The level of γ-kafirin was particularly higher in the vitreous endosperm portions of these kernels. The ratio of γ-kafirin to the α-kafirin was, however, higher for the floury portions of soft kernels. Tissue print immunoblotting revealed that the β- and γ-kafirins were concentrated in the central floury endosperm portions of soft kernels, whilst α-kafirin was distributed more uniformly throughout the endosperm. In contrast, all three kafirins were distributed uniformly throughout the endosperms of hard kernels. The data indicate that the content, as well as the distribution, of kafirins within the kernel is different in grains varying in endosperm hardness.     

Evaluation of malting barley quality using exploratory data analysis. II. The use of kernel hardness and image analysis as screening methods

This paper presents an exploratory investigation of the use of image analysis and hardness analysis of barley kernels for characterisation and prediction of malting quality. A sample set of fifty barley samples representing 15 spring barley and 10 winter barley varieties grown at two locations in Denmark was used. The samples were micro-malted and mashed and analysed for 13 quality parameters according to the official methods of the European Brewery Convention. A sub-sample of the barley samples was analysed on two different single kernel instruments: (1) Foss Tecator GrainCheck was applied for non-destructive recording of single kernel size and shape (width, length, roundness, area, volume and total light reflectance) and (2) Perten Single Kernel Characterization System 4100 was applied for single kernel hardness and weight determinations. The eight variables from these single seed analyses have been used in two different ways, either as means and standard deviations, or as appended histogram spectra representing 250 kernels from each bulk sample. By the two methods, it has been possible to obtain reasonable Partial Least Squares Regression (PLSR) models for the structural and physical part of the malting quality complex associated to malt modification, but it was as expected impossible to predict the biochemical parameters associated with nitrogen chemistry and enzymatic power. The best model was achieved for (1→3, 1→4)-β-D-glucan in barley. The hardness of the barley kernels is by far the most important variable for describing malting performance. The additional use of the morphological data as acquired by fast non-destructive image analysis, however, also reveals some malting quality information by improving the calibration models based on hardness alone. The brightness of the kernels is by far the most important GrainCheck variable but also kernel size and shape is associated to malting performance. In general, the utilisation of the single kernel readings (used as histogram spectra), compared to sample mean and standard deviation, did not provide additional information for an improved prediction of the malting quality parameters.     

Identification of Damaged Kernels in Wheat using a Colour Machine Vision System

A colour machine vision system was used for identification of healthy and six types of damaged kernels [broken, grass-green/green-frosted, black-point/smudged (selected from grain samples), mildewed, heated and bin-/fire-burnt (created in the laboratory)] of Canada Western Red Spring (CWRS) wheat. A software package was developed to extract various morphological and colour features from the images of both healthy and damaged kernels. Different feature models, morphological, colour and combined (morphological and colour), were evaluated for the identification analysis using the SAS procedures, STEPDISC and DISCRIM. Both parametric and non-parametric statistical classification methods were evaluated with the selected feature models. Colour features proved to be efficient for the identification of healthy and damaged kernels, while combining morphological with colour features improved the identification accuracy. Using a non-parametric classifier with a selected combined model of 24 colour and 4 morphological features, the average identification accuracies were: 93% (healthy), 90% (broken), 99% (grass-green/green-frosted), 99% (black-point/smudged), 99% (mildewed), 98% (heated), and 100% (bin-/fire-burnt), when trained and tested with three different training and testing data sets.     

Vitreosity,its stability and relationship to protein content in durum wheat

High quality requirements are set on durum wheat (Triticum durum) from semolina mills and pasta producers. For the production of semolina and pasta with good cooking quality, high grain protein content and vitreosity is required. The dependency of vitreosity on protein content as well as its stability under the influence of humidity was not well investigated up to now. We (1) compared two methods to determine vitreosity, (2) investigated the relationship between vitreosity and protein content, (3) developed a method to analyze vitreosity under humidity, and (4) examined the relationship between protein content and agronomical as well as quality traits in durum wheat. The results showed that the formation of vitreous kernels greatly depends on the protein content. To evaluate the stability of vitreosity under the influence of humidity a new method was elaborated and employed to assess the durum germplasm under study. This revealed that vitreosity of a durum wheat variety depends on the potential to form vitreous kernels but also to maintain this vitreosity under the influence of humidity. Our results further show that protein content is a central trait in durum wheat that strongly influences important traits like grain yield, vitreosity, and b-value.     

Wheat class identification using monochrome images

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.     

Reliability of InGaAs focal plane array imaging of wheat germination at early stages

To assist our Kansas breeding program, we have developed nondestructive methods to test new lines of wheat for resistance to premature germination. The high sensitivity of subsurface imaging, compared with visual detection, α-amylase determination, or viscosity testing, permits germination detection at early stages. This report is concerned with detection reliability via chemical imaging of intact wheat kernels at early stages of germination by using an InGaAs focal plane detector array in the 1100–1700 nm range. Ninety kernels from each of six different cultivars, including HRW and HWW wheat, were exposed to moist conditions for 6, 12, 24, 36, and 48 h. Images of each 90 kernel group were examined, and images of those kernels exposed to moisture for 36 h were compared with images of kernels treated for 3 h as a control. Images of each were classified as sprouted or unsprouted. Criteria for classification included images of log 1/R at select wavelengths or images of select factors resulting from principle component analysis (PCA) treatment of reflectance intensity data from each pixel. Sprouted kernels determined by PCA factors 1 and 4 from 90 kernels tested in a 36-h moisture exposure numbered 87, 85, 80, 74, 70, 48 for six cultivars tested. Cultivar KS-2174 was shown to be distinctly more resistant to germination than the other cultivars. When KS-2174 was compared with Betty wheat, for all exposure times, Betty had approximately 45% more germinated kernels.     

Kernel vitreousness and protein content: Relationship,interaction and synergistic effects on durum wheat quality

Four sets of durum samples were used in this study to further understand the interrelationships among hard vitreous kernels (HVK), protein content, and pigment concentration, with a focus on the interaction and synergistic effects of protein content and vitreousness on durum quality. HVK level increases with higher protein content in the range of 9.5–12.5%, but this relationship is less evident in durum samples with high protein content (12.5–14.5%). Both protein content and kernel vitreousness can significantly affect durum milling quality. White starchy kernels (WSK) in low protein durum have a very detrimental impact on milling and pasta processing quality, but high protein content can mitigate the adverse impact of WSK on durum quality. Although protein content plays a dominant role, higher HVK might contribute positively to pasta firmness. There was no significant difference in yellow pigment content between HVK and WSK. However, pigment loss from semolina to dough was higher for WSK than HVK. Despite the difference in protein content, HVK and WSK have little difference in gluten strength. The monomeric protein was preferentially accumulated in HVK. The glutenin proteins of HVK and WSK were similar in the ratios of 1Bx/1By and HMW/LMW-GS.     

Water Vapor Diffusivity in Vitreous and Mealy Wheat Endosperm

The water vapor diffusivities of vitreous and mealy wheat endosperms were measured using diffusion cells containing wax-scaled cylinders of endosperm. The diffusion cells were incubated in a chamber that was maintained at a constant temperature and relative humidity. A water vapor gradient was created by filling each diffusion cell with desiccant. A steady-state condition was achieved within 24 h. The mean diffusivities for the vitreous endosperm samples of cvs Arizona and Len were 2·0 × 10^-8 cm²/s. The diffusivities for mealy samples of cvs Logan and Titan were 3·5 and 9·1 × 10^-8 cm²/2, respectively. The liquid water diffusivities of whole caryopses were similar for five vitreous and five mealy wheat samples tested. No change was detected in the diffusion mechanism for any of the samples in the temperature range tested. The energy of activation (E) was significantly (P < 0·05) higher for endosperm for cv. Len, a vitreous sample, than for cv. Titan, a mealy sample.     

Measurement and fissuring of rice kernels during quasi-moisture sorption by image analysis

Fissuring associated with kernel elongation was evident in a moisture-adsorbing environment at 15 °C of relative humidity 88.72±0.28% with moisture content from 6% (d.b.) to 16.3% (d.b.). The average length and width of 800 kernels was measured during the sorption test using a Cervitec Grain Inspector equipped with two digital cameras. Samples were exposed to both a moisture-adsorbing and desorbing environments at 15 °C with relative humidities of 23.4±0.53%, 55.9±0.27%, 70.98±0.28% and 88.72±0.28%. On image analysis, the kernels changed in size from 5.13 to 4.84 mm in length and 2.9 to 2.73 mm in width. Both length and width changed as expected during moisture-adsorbing and moisture-desorbing tests. Fissured kernel percent was related to changes in the average length of the kernels during the moisture-adsorbing tests. The average length and fissured kernel percent of rice kernels measured by image analysis may be considered as a reliable indicator of changes in their dimensions during quasi-static moisture sorption processes.     

小麦黑胚病籽粒黑变物质的理化性质分析

黑胚病是危害小麦生产的重要籽粒病害。为了解小麦黑胚病症状产生的原因,用碱溶酸沉的方法提取感黑胚病籽粒黑变部位的物质(黑变物质)与健康籽粒相同部位提取物(BPF)以及致病菌麦根腐平脐蠕孢(Bipolaris sorokiniana)的胞内黑色素与胞外黑色素,并对其进行紫外-可见吸收光谱、红外光谱扫描以及光、热稳定性等理化性质分析。结果显示:(1)病粒黑变物质的紫外吸收峰(235 nm)与B.sorokiniana的胞外、胞内黑色素的紫外吸收峰不一致(210 nm),而与BPF的紫外吸收峰相同;(2)病粒黑变物质与BPF的红外光谱一致,而与B.sorokiniana的胞外、胞内黑色素不一致;(3)病粒黑变物质与BPF的光、热稳定性较差,而来源于B.sorokiniana的胞内、胞外黑色素相对稳定。综上所述,黑胚病籽粒的黑变物质与致病菌B.sorokiniana的黑色素的理化性质有较大差异,而与健康粒相同部位提取物的理化性质类似,推测该物质是由小麦籽粒本身合成,而不是来源于病原真菌。     

The composition of floury and vitreous endosperm affects starch digestibility kinetics of the whole maize kernel

Maize grain starch is the major energy source in animal nutrition, and its high digestion and utilization largely depend on endosperm traits and the structure of the starch-lipoprotein matrix. The aim of this work was to determine floury and vitreous endosperm traits and its relation to starch digestibility rate. In total, kernels of 30 hybrids were manually dissected, and amylose, total zein and starch and non-starch lipids were determined in both vitreous and floury endosperm. Starch digestibility of the whole kernel was determined based on glucose released during a two-step in vitro pig model of enzymatic digestion, and starch digestibility rate was calculated according to the first-order kinetics. The vitreous endosperm of tested hybrids had higher contents of amylose (204.6 vs 190.4 g/kg), zein (63.2 vs 40.4 k/kg) and starch lipids (5.6 vs 4.9 g/kg), and lower content of non-starch lipids (7.3 vs 9.6 g/kg) than floury endosperm. Digestibility coefficients varied among hybrids, and starch digestibility rate varied from 0.73 to 1.63 1/h. Lipids in both vitreous and floury endosperm negatively correlated with the most of digestion coefficients, whereas zein correlated in vitreous and amylose in the floury endosperm (P < 0.05). Starch digestibility rate negatively correlated with all traits, except amylose content in vitreous endosperm. As a result, a linear regression model with four variables including contents of zein and starch lipids in vitreous and zein and amylose in floury endosperm can predict more than 65% variability of starch digestibility rate of tested hybrids.