Identification of multiple grain image samples from tray

This paper presents a methodology to identify the different grain types from image samples of tray containing multiple grains using colour and textural features. The multiple grain images are segmented into individual grain images. From these images, eighteen colour and twenty‐four textural features are obtained. A neural network model is implemented for identification of bulk food grains. Five different types of grains namely, alasandi, green gram, metagi, red gram and wheat commonly used in Indian food preparations are considered in this work. The maximum and minimum food grain identification accuracies observed in this work are 94% and 80% for wheat and alasandi, respectively. The work finds application in development of machine vision system for grain identification, classification and grading.     

Detection of internal wheat seed infestation by Rhyzopertha dominica using X-ray imaging

Standardization for grain grades has been established in most countries to maintain the quality of a crop until it reaches consumers. Different methods have been investigated for their potential to detect insect infestations in grain destined for domestic and export markets. The potential of detecting infestations caused by Rhyzopertha dominica in wheat kernels using a real-time soft X-ray method was determined in this study. Artificially infested wheat kernels were incubated at 30°C and 70% relative humidity and X-rayed sequentially for larval, pupal, and adult stages of R. dominica. Algorithms were used to extract histogram features, textural features, and histogram and shape moments from the X-ray images of wheat kernels. A backpropagation neural network (BPNN) and statistical classifiers were used to identify uninfested and infested kernels using the 57 extracted features. The BPNN correctly identified all uninfested and infested kernels and more than 99% of kernels infested by R. dominica larvae. The classification accuracies determined by the BPNN were higher using all 57 features than when using the histogram and textural features separately. The BPNN performed better than the parametric and non-parametric classifiers in the identification of uninfested and infested kernels by different stages of R. dominica.     

Estimating mycotoxin contents of Fusarium-damaged winter wheat kernels

Winter wheat (Triticum aestivum L., cultivars Ritmo and Dekan) grain was sampled in Northern Germany between 2001 and 2006. Kernels damaged by fungi of the genus Fusarium were separated from sound grain by visual assessment. Samples containing 0%, 20%, 40%, 60%, 80% and 100% of Fusarium-damaged kernels were compiled and analyzed for the Fusarium type B trichothecenes deoxynivalenol (DON, 2001-2006), nivalenol (NIV, 2006), 3-acetyl-deoxynivalenol (3AcDON, 2006) and 15-acetyl-deoxynivalenol (15AcDON, 2006). The relationship between mycotoxin contents and the percentage of Fusarium-damaged kernels was calculated for each lot of grain. Apart from one exception, relationships between the percentage of Fusarium-damaged kernels and NIV, 3AcDON or 15AcDON were non-significant. In contrast, close relationships between the percentage of Fusarium-damaged kernels and the DON content were observed (r(2)=0.93-0.99). The y-axis intercepts were not significantly different from zero, but the DON content of the damaged kernels varied by a factor of 11.59 between years and by a factor of 1.87 between cultivars. Fusarium-damaged kernels contained between 0.21 and 2.39 microg DON kernel(-1). The overall average DON content of a Fusarium-damaged wheat kernel was 1.29 +/- 0.11 microg. The DON content of diseased kernels was affected by environment and wheat genotype but not by genotype x environment interaction. On average, Fusarium-damaged kernels contained 9.7-fold more DON than 15AcDON, 19.5-fold more DON than NIV, and 26.9-fold more DON than 3AcDON. 3AcDON and 15AcDON contents per wheat kernel were not significantly different between cultivars. On average, 4.27% of Fusarium-damaged kernels were sufficient to reach the 1.25 mg DON kg(-1) grain limit for unprocessed cereals in the EU. Given the low percentages of Fusarium-damaged kernels that are equivalent to current legal DON limits, grading accuracies >96% would be needed when using automatic grading systems for separating sound from damaged kernels.     

Identification of wheat classes at different moisture levels using near-infrared hyperspectral images of bulk samples

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.     

双扩散传热传质模型及在横向谷冷通风中应用

本文首先建立了单颗粒小麦内部水分迁移模型,基于有限元的方法数值模拟了颗粒内部水分变化规律,通过回归数值模拟数据,得到了颗粒平均水分模型和平均水分变化的干燥(或吸湿)速率模型。在此基础上推导出了谷物颗粒堆积床双扩散传热传质模型,并采用有限元的方法数值模拟分析了就仓横向(水平)谷冷通风时仓储粮堆内部热湿耦合传递规律。通过比较数值模拟和实验测定数据,验证了所建立的模型的合理性。分析了横向谷冷通风时粮粒温度和水分以及粮粒周围空气温度的变化规律,探讨了横向谷冷通风时粮堆内部降温效果。     

含水量对小麦麦仁膨化品质的影响

分析了小麦麦仁的含水量对麦仁膨化品质的影响,为小麦麦仁的精深加工提供科学依据。将小麦麦仁含水量调节到8%,10%,12%1,4%,16%,18%和20%7个水平,测定麦仁膨化品质。结果表明,在小麦麦仁品质和膨化压力等因素一致的条件下,小麦麦仁含水量对麦仁膨化品质的影响非常明显。最大的膨胀体积,鳞片大小和最小的未膨胀率在麦仁分含水量为16%取得。另外,膨化后的麦仁粉峰值黏度,谷值黏度,末值黏度,衰减值和回生值都比未膨化麦仁粉低,膨化后的麦仁粉凝胶硬度和咀嚼性比原麦仁粉降低。     

The conditioning of wheat. An investigation into the conditioning requirements of Canadian western red spring no. 1

Canadian Western Red Springs (CWRS) No. 1 wheat has been conditioned to 16 and 17% moisture content by 2 and 4% water additions in each case. Lying times in the range 4–48 h before milling have been investigated. The milling performance of the conditioned wheat has been assessed in terms of extraction rate, colour and flour moisture content. Lying times in excess of 12 h were found to have no influence on extraction rate, colour and flour moisture content. For lying times less than 12 h the extraction rate was affected by the level of added water. When CWRS No. 1 wheat was milled at moisture contents in the range 14–18% using a fixed lying time, extraction rate, flour colour and ash fell with increasing wheat moisture content while flour moisture content increased.     

Dropping macadamia nuts‐in‐shell reduces kernel roasting quality

BACKGROUND: Macadamia nuts (‘nuts‐in‐shell’) are subjected to many impacts from dropping during postharvest handling, resulting in damage to the raw kernel. The effect of dropping on roasted kernel quality is unknown. Macadamia nuts‐in‐shell were dropped in various combinations of moisture content, number of drops and receiving surface in three experiments. After dropping, samples from each treatment and undropped controls were dry oven‐roasted for 20 min at 130 °C, and kernels were assessed for colour, mottled colour and surface damage. RESULTS: Dropping nuts‐in‐shell onto a bed of nuts‐in‐shell at 3% moisture content or 20% moisture content increased the percentage of dark roasted kernels. Kernels from nuts dropped first at 20%, then 10% moisture content, onto a metal plate had increased mottled colour. Dropping nuts‐in‐shell at 3% moisture content onto nuts‐in‐shell significantly increased surface damage. Similarly, surface damage increased for kernels dropped onto a metal plate at 20%, then at 10% moisture content. CONCLUSION: Postharvest dropping of macadamia nuts‐in‐shell causes concealed cellular damage to kernels, the effects not evident until roasting. This damage provides the reagents needed for non‐enzymatic browning reactions. Improvements in handling, such as reducing the number of drops and improving handling equipment, will reduce cellular damage and after‐roast darkening. Copyright © 2010 Society of Chemical Industry     

Effects of insect-infested kernels on bulk flow properties of wheat

Insect damage to kernels during storage affects the processing quality of wheat and as well as bulk wheat properties, which in turn causes hopper flow problems such as funnel flow, ratholing, arching, or flushing. This study focused on kernel damage by Rhyzopertha Dominica F. (lesser grain borer), one of the most commonly found insects in wheat, and resulting changes in physical properties, such as bulk density, tapped density, true density, and angle of repose, and in flow properties, such as basic flowability energy, stability, aeration, compressibility, and permeability. Bulk and tapped densities of sound hard red winter wheat kernels and infested wheat kernels were about 720 kg/m³ and 775 kg/m³, respectively. Compressibility index (CI), Hausner ratio (HR), and angle of repose of infested wheat kernels were higher than for sound hard red winter wheat kernels, indicating decreased flowability. Grain dust and insect-infested kernels could form localized compacted areas within the grain bins, resulting in uneven flow during discharge. Results from this study indicate that the presence of insect-infested kernels could lead to arching and bridge formation in grain bins, thus affecting the flow characteristics of bulk wheat.     

基于差分图像边界距离的粮粒孔洞自动检测

粮粒孔洞的自动检测是近红外高光谱图像技术检测粮粒内部害虫中的一个关键问题。提出基于差分图像边界距离的粮粒孔洞自动检测方法,该方法通过求取粮粒(内部)轮廓与阈值分割后二值图像的差分,若差分图像中的目标与粮粒边界的最远距离大于某个阈值时,则该目标应判别为边界(内部)孔洞。用米象的幼虫、蛹和成虫3个侵染阶段粮粒的900帧近红外图像进行训练,用450帧近红外图像进行检验,结果表明该方法不仅可以判断粮粒是否存在孔洞,还能检测出孔洞的数量及形态,其中边界孔洞和内部孔洞的识别率分别为97.33%和95.56%,证实了基于差分图像边界距离的粮粒孔洞检测方法是可行的。     

油茶籽鲜储过程中水分迁移与质构特性的变化

为明确油茶籽采后鲜储过程中水分迁移及质构变化情况,本实验以黔产油茶籽为实验原料,运用低场核磁共振(low field nuclear magnetic resonance,LF-NMR)、核磁共振成像(magnetic resonance imaging,MRI)技术探究鲜储过程中水分状态和分布规律,采用质地剖面分析(texture profile analysis,TPA)监测油茶籽鲜储过程中的质构特性变化,并进行相关性分析。结果表明:油茶籽在储藏期间水分含量不断下降,黔玉1号油茶籽水分含量由33.86%±3.03%(0 d)降至8.64%±0.24%(56 d),湘林210号油茶籽水分含量由53.03%±3.36%(0 d)降至10.73%±0.25%(56 d);种仁水分含量下降速率高于油茶籽;不易流动水在油茶籽中占比最高,储藏56 d时黔玉1号降至62.89%,湘林210号降至60.64%。新鲜油茶籽氢质子密度成像图光亮,随时间延长,局部水分流失较快,图像逐渐接近背景色。储藏期间,油茶籽破裂力、硬度不断下降,种仁破裂力、硬度曲折变化;种仁弹性逐渐丧失;黔玉1号种仁内聚性变化较小...     

A comparison of the functional characteristics of wheat stored as grain with wheat stored in spike form

In this study, wheat grain and wheat spike with 12%, 14% and 16% moisture content were stored at 10, 20 and 30 °C for 0, 3, 6 and 9 months. After storage, wheat samples were investigated for hectolitre weight, gluten content, Zeleny sedimentation volume, enzyme activity, acidity, phytic acid and L colour value. Storage of wheat at different storage forms (spike and grain) and storage conditions showed considerable changes in grain quality. In general, the storage period of 3 months positively affected wheat quality. However, hectolitre weight, gluten, Zeleny sedimentation, enzyme activity, acidity and colour of wheat got worse at storage periods beyond 3 months. Hectolitre weight, wet and dry gluten, Zeleny sedimentation, phytic acid content and L Colour value of wheat stored in both spike and grain form significantly decreased during storage. However, the increase in grain moisture content, storage time and temperature resulted in significant increase in total titratable acidity and falling number values of wheat. Falling Number and phytic acid values of wheat stored in spike form were generally lower than wheat stored in grain form. Storage in spike form had a positive effect on especially wet gluten content of wheat stored at non-optimal storage conditions such as high grain moisture content and high temperature. Wet gluten of wheat stored in spike form was higher than that of wheat stored grain form after storage at 30 °C for 6 and 9 months. Wheat stored in spike form is more resistant than wheat stored in grain form against adverse storage conditions such as high moisture content and temperature and longer storage time.     

优质籼米地下仓储藏过程中品质变化

为探索成品粮绿色、安全的储藏方法,在南方高温高湿地区,利用地下仓储藏黄花粘和籼优998两种优质籼米,定期检测水分、碎米总量、小碎米率、垩白粒率、黄粒米、直链淀粉、品尝评分值、色泽气味、不完善率、害虫虫口密度等指标。实验结果显示,实验期间,大米水分呈小幅下降;2种大米不生虫时间长达13个月;小碎米率、色泽气味等指标无显著性变化;碎米总量、黄粒米、不完善粒含量小幅增加,直链淀粉含量逐渐增加,但均未导致大米质量等级下降;垩白粒率显著增加,品尝评分值显著下降,且均造成大米质量定等下降。根据国家标准《大米》的规定,籼优998于储藏3个月后,质量定等由二级降为三级,储藏6个月后,已不属于优质籼米,黄花粘于实验结束后仍为优质籼米三级。说明地下仓具有良好的保湿、保鲜和害虫抑制效果,优质籼米耐储性能与储藏条件和品种均相关。     

Using a Short Wavelength Infrared (SWIR) hyperspectral imaging system to predict alpha amylase activity in individual Canadian western wheat kernels

Sprout damage (pre-harvest germination) in wheat results in highly deleterious effects on end-product quality. Alpha-amylase, the pre-dominant enzyme in the early stage of sprouting has the most damaging effect. This paper introduces a new method using a SWIR hyperspectral imaging system (1000–2500 nm) to predict the α-amylase activity of individual wheat kernels. Two classes of Canadian wheat, Canada Western Red Spring (CWRS) and Canada Western Amber Durum (CWAD), with samples of differing degrees of sprout damage were investigated. Individual kernels were first imaged with the hyperspectral imaging system and then the α-amylase activity of each kernel was determined analytically. Individual kernel α-amylase activity prediction was significant (R ² 0.54 and 0.73) for CWAD and CWRS, respectively using Partial Least Square regression on the hyperspectral data. A classification method is proposed to separate CWRS kernels with high α-amylase activity level from those with low α-amylase activity giving an accuracy of above 80%. This work shows that hyper/multi-spectral imaging techniques can be used for rapidly predicting the α-amylase activity of individual kernels, detecting sprouting at early stage.     

基于软X射线与低场核磁检测小麦隐蔽性害虫玉米象

针对小麦内部隐蔽性害虫玉米象(Sitophilus zeamais)难以检出的问题,本文将高清软X射线与低场核磁两种检测技术相结合,通过高清软X射线拍摄的图片,观察玉米象在小麦内部的整个生长周期,提取图片纹理特征,用线性判别分析(linear discriminant analysis,LDA)与二次判别分析(quadratic discriminant analysis,QDA)算法进行分类判别,并对被不同虫态玉米象感染的小麦进行低场核磁检测。研究结果表明:在12%水分含量小麦中,玉米象的生长周期大约为36 d,LDA与QDA模型对未感染小麦与不同感染阶段小麦进行单独分类判别时,准确率都达到了95%以上,对卵期以及幼虫具有较高的分类准确率。根据小麦被玉米象感染后特征峰值比例P2b与P22的变化,可以用来作为小麦是否受玉米象感染定性判断的依据。     

Application of Image Texture Analysis for Evaluation of X-Ray Images of Fungal-Infected Maize Kernels

The feasibility of image texture analysis to evaluate X-ray images of fungal-infected maize kernels was investigated. X-ray images of maize kernels infected with Fusarium verticillioides and control kernels were acquired using high-resolution X-ray micro-computed tomography. After image acquisition and pre-processing, several algorithms were developed to extract image textural features from selected two-dimensional (2D) images of the kernels. Four first-order statistics (mean, standard deviation, kurtosis and skewness) and four grey level co-occurrence matrix (GLCM) features (correlation, energy, homogeneity and contrast) were extracted from the side, front and top views of each kernel and used as inputs for principal component analysis (PCA). The first-order statistical image features gave a better separation of the control from infected kernels on day 8 post-inoculation. Classification models were developed using partial least squares discriminant analysis (PLS-DA), and accuracies of 67 and 73% were achieved using first-order statistical features and GLCM extracted features, respectively. This work provides information on the possible application of image texture as method for analysing X-ray images.     

基于高光谱图像光谱变量和颜色特征的霉变玉米籽粒识别

目的:准确识别霉变玉米籽粒。方法:基于高光谱图像光谱变量和颜色特征建立霉变玉米籽粒识别的新方法。先对玉米籽粒图像进行图像分割和光谱变量、颜色特征提取,并根据颜色特征生成颜色直方图;将光谱变量和颜色直方图特征组成特征集合;通过距离函数对特征集合中所有特征的分析确定霉变玉米籽粒所属类别。结果:所提方法对霉变玉米籽粒类别的最大平均识别偏差为1.12,最佳平均识别准确率为97.59%;与基于高光谱图像+随机蛙跳+极限学习机的方法、基于高光谱图像+稀疏自动编码器+卷积神经网络的方法、基于高光谱图像+蚁群优化+BP神经网络的方法相比,研究所提方法对霉变玉米籽粒类别的识别准确率明显提高。结论:该方法可实现被测玉米籽粒样品是否霉变以及霉变程度的准确判断。     

Estimating deoxynivalenol contents of wheat samples containing different levels of Fusarium-damaged kernels by diffuse reflectance spectrometry and partial least square regression

Fusarium head blight is a fungal disease causing yield losses and mycotoxin contamination in wheat and other cereals. Wheat kernels (cultivar Ritmo) were sampled in 2001, 2002, 2003, and 2006 and Fusarium-damaged kernels were separated from sound grain based on visual assessment. Subsequently, grain lots containing 0, 20, 40, 60, 80, and 100% of damaged kernels were compiled. Each lot was split and the spectrometric reflectance (wavelengths 350-2500 nm) was measured using subgroup one, while the concentration of the mycotoxin deoxynivalenol (DON) was determined by high-performance liquid chromatography in subgroup two. DON concentrations in batches classified as sound were not significantly different from 0. Estimating DON contents from the percentage of Fusarium-damaged kernels was impeded by vast variability, resulting in a coefficient of determination of 0.49. Using spectrometric data subjected to partial least square regression allowed estimating DON contents with higher accuracy, in particular at elevated percentages of damaged kernels. The coefficient of determination was 0.84 for the relationship between DON contents estimated based on spectrometric data and the DON contents measured. The intercept of a regression line fitted through a plot of estimated versus measured DON contents was 0.89 ± 3.61 mg/kg. Since intercept + standard error was larger than the actual legal limit (1.25 mg DON per kg dry grain in the European Union), the spectrometric procedure was still not precise enough to allow a reliable separation of grain samples with DON contents below 1.25 mg/kg from samples with DON contents above the limit. However, spectrometric data also allowed estimating the DON content of the average damaged kernel within a given lot composed of sound and damaged kernels, which is probably the reason for the reduction of the fraction of unexplained variance by 35% compared to the visual approach and illustrates that spectrometric approaches can make a contribution to reducing DON contents of wheat grain.     

Wheat Class Identification Using Thermal Imaging

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.     

Influence of Variety,Moisture Content,Kernel Size and Applied Current Frequency on the Electric Properties of Wheat Grain

The work aimed at determination of electric properties of wheat grain in dependence on its variety, moisture, geometrical features of kernels and applied current frequency. Wheat grain of 4 Polish winter varieties: Korweta, Juma, Mikon and Kobra from harvest 2001 were used as the material for study. Grain was sized into 3 fractions: (1) > 2,8mm, (2) 2,5–2,8mm, and (3) 2,2–2,5mm. Basic geometrical features were determined for not sorted grain (control sample) and its three fractions by the use of digital image analysis. Electric properties of grain (at 11% and 15% moisture content) have been performed with the Hewlett Packard 4263B meter. Measurements of impedance, resistance, admittance, conductance, as well as equivalent parallel capacitance and equivalent series capacitance were made. Obtained results were subjected to statistical analysis with the use of Statistica? programme. Changes in electric properties of grain significantly depended on all of studied factors. Most of all significant correlations appeared between geometrical features and studied electric properties of grain of 15% moisture. Statistical analysis of the results proved significant linear correlations between electric properties of kernels and their length, perimeter and circularity coefficient R_C2 at higher measurement frequencies.