基于图像处理和神经网络的小麦不完善粒识别方法研究

为了实现图像处理技术对小麦不完善粒的准确快速识别,研究了一种基于小麦不完善粒图像特征和BP神经网络的不完善粒识别方法。采集小麦不完善粒图像,对图像进行中值滤波、形态学运算、图像分割等处理后,针对每个小麦籽粒,提取其形态、颜色和纹理共3大类54个特征参数,采用主成分分析法提取8个主成分得分向量作为模式识别的输入,建立BP神经网络模型,实现对小麦不完善粒的检测识别。结果表明,该模型对完善粒、破损粒、病斑粒、生芽粒和虫蚀粒的判别正确率分别为93%、98%、100%、90%和85%,平均判别正确率达到93%,可有效对小麦不完善粒进行检测识别。     

基于CNN神经网络的小麦不完善粒高光谱检测

利用高光谱成像技术对小麦不完善粒进行无损检测。以932个小麦为样本,其中正常粒样本486个、破损粒样本170个、虫蚀粒样本149个及黑胚粒样本127个为研究对象,通过高光谱图像采集系统采集样本的光谱信息,然后从每个样本的116个波段中选取30个波段,建立基于深度学习的卷积神经网络(convolutional neural networks,CNN)模型。实验中的CNN采用2个卷积层,第1层采用大小为3×3的32个卷积核,第2层采用大小为5×5的64个卷积核,池化层采用最大池,激活函数采用修正线性单元,为避免过拟合,在全连接层后面接入dropout层,参数设置为0.5,其他卷积参数均为默认值,得到校正集总识别率为100.00%,测试集总识别率为99.98%。最后,以支持向量机(support vector machine,SVM)为基线模型进行对比,从116个波段中选取90个波段进行建模,测试集总识别率为94.73%。通过实验对比可以看出,CNN模型比SVM模型识别率高。研究表明CNN模型能够实现对小麦不完善粒的准确、快速、无损检测。     

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。     

Thermal imaging to detect infestation by Cryptolestes ferrugineus inside wheat kernels

Canada's zero tolerance for live insects in grain received from farmers, and shipped to domestic and export buyers, has necessitated the development of an accurate insect detection method. An infrared thermal imaging system was developed to detect infestation by six developmental stages (four larval instars, pupae and adults) of Cryptolestes ferrugineus under the seed coat on the germ of the wheat kernels. The artificially infested wheat kernels were removed from the incubation room (30 °C), refrigerated (5 °C) for 60 s, maintained at ambient conditions for 20 s, and imaged using a thermal camera to identify each developmental stage (n=283). The means of the highest 5% and 10% of all temperature values on the surface of the grain were significantly higher (=0.05) for grains having young larvae inside and lower for grains having pupae inside. Temperature distribution on the surface of the infested kernels with different stages of C. ferrugineus was highly correlated with the respiration rate of each developmental stage (r=0.83–0.91). The overall classification accuracy for a quadratic function was 83.5% and 77.7% for infested and sound kernels, respectively, and for a linear function, it was 77.6% and 83.0% for infested and sound kernels, respectively, in pairwise discriminations. Thermal imaging has the potential to identify whether the grain is infested or not, but is less effective in identifying which developmental stage is present.     

Identification of Broken Rice Kernels Using Image Analysis Techniques Combined with Velocity Representation Method

The percentage of whole kernels remaining after milling is one of the most important physical characteristics of rice quality. A method based on flatbed scanning and image analysis was developed for the identification of broken rice kernels. Velocity representation method was developed for pattern recognition based on the contour characteristics of the rice kernels. The similarity of the boundary features of the image of rice kernel was measured by similarity coefficient, which was used to identify the broken rice kernel by comparing with threshold. High recognition rates for three rice varieties were reached by this method with 96.7% for Thailand rice, 98.73% for Pearl rice, and 97.14% for Changlixiang rice, respectively, and the recognition rate could be improved by the adjustment of the similarity coefficient threshold. Because the comprehensive boundary features were the basis for the classification, this method could be more accurate compared to other methods using the single dimension feature.     

The breakage susceptibility of raw and parboiled rice: A review

Rice (Oryza sativa L.) is one of the most important cereals in the world. Before it is consumed, it is common to remove the hull, bran and germ from the rough rice kernel which is either parboiled or not. During such processing, rice kernels are subjected to mechanical stresses which cause some rice grains to break. A main challenge of the rice industry is to minimize the quantities of broken rice. We here review the factors impacting the breakage susceptibility of rice kernels. Their tendency to break is primarily determined by fissures, chalkiness, immaturity and rice kernel dimensions, properties which are both cultivar and rice grain history dependent. The intensity of processing of any given rice feedstock determines the actual level of broken rice kernels. If performed properly, parboiling, a three-step hydrothermal treatment consisting of soaking, heating and drying of rough rice, substantially reduces the level of broken kernels.     

斯里兰卡红碎茶与滇红碎茶主要香气成分比较分析

目的 比较分析斯里兰卡红碎茶与滇红碎茶香气成分及含量的差异性,探明区分两地红碎茶差异的主要香气成分。方法 采用GC-MS法测定斯里兰卡红碎茶及滇红碎茶样品的香气成分,通过主成分分析法比较斯里兰卡红碎茶与滇红碎茶香气成分特征差异性,根据第一主成分贡献值大小,筛选出决定区分两地红碎茶差异的主要香气成分。结果 两地红碎茶样品共分析到香气化合物40种,以醛类、酮类、酯类、醇类为主,香气成分及含量相近,但略有差异。其中反-2-己烯醛、苯乙醇、香叶醇、壬醛香气成分滇红碎茶明显高于斯里兰卡红碎茶（高出80%以上）,分别高出110.00%、108.73%、93.64%、84.91%,二氢猕猴桃内酯、吲哚香气成分滇红碎茶明显低于斯里兰卡红碎茶（50%以上）,分别低至56.19%、52.38%。其中烯类香气成分滇红碎茶比斯里兰卡红碎茶高出最为明显,高出53.53%,吲哚类香气成分滇红碎茶比斯里兰卡红碎茶低出最为明显,低出52.48%。两类红碎茶中,水杨酸甲酯占比最高,在斯里兰卡红碎茶中最高可达26.60%,在滇红碎茶中最高可达29.79%。斯里兰卡红碎茶和滇红碎茶中,酯类和醇类含量最高,酯类在斯里兰卡红碎茶中最高可达34.38%,在滇红碎茶中最高可达35.69%,醇类在斯里兰卡红碎茶中最高可达32.48%,在滇红碎茶中最高可达34.84%。各香气成分中二氢猕猴桃内酯（C15）、反-戊酸-2-己烯酯（C29）最能代表斯里兰卡红碎茶香气特征,a-萜品醇（C30）、橙花醇（C40）最能代表滇红碎茶香气成分特征。结论 对香气成分进行分析,可明显区分斯里兰卡红碎茶与滇红碎茶特征。     

Preparation of rice analogues using extrusion technology

Rice is considered as staple food in many parts of the world. An issue of concern is the breakage of rice kernels in milling processes, and these broken kernels are not generally accepted by consumers. These broken kernels can be mixed with some desired additives to improve their quality and extruded for the preparation of reconstituted rice kernels or rice analogues. Various studies have been conducted for the preparation of the rice analogues in the past few decades, and recently attempts have been made to fortify these analogues with protein, certain vitamins and minerals. The main features such as colour, shape, size, texture, and cooking characteristics and cooking time of these rice‐like grains can be tailored to the requirements of specific applications by modification of the extrusion parameters. Various organisations, such as Wuxi NutriRice Co. (DSM/Buhler) and China National Cereals, Oils and Foodstuffs Corporate (COFCO), Superlative Snacks Inc., Vigui and PATH, have utilised this technique to prepare fortified and reconstituted rice. Studies have shown that it is possible to improve the nutritional quality of rice by fortified rice analogues. This article reviews research results of the many approaches to the formation of fortified rice analogues by extrusion‐based technologies.     

Quality changes in macadamia kernel between harvest and farm-gate

BACKGROUND: Macadamia integrifolia, Macadamia tetraphylla and their hybrids are cultivated for their edible kernels. After harvest, nuts‐in‐shell are partially dried on‐farm and sorted to eliminate poor‐quality kernels before consignment to a processor. During these operations, kernel quality may be lost. In this study, macadamia nuts‐in‐shell were sampled at five points of an on‐farm postharvest handling chain from dehusking to the final storage silo to assess quality loss prior to consignment. Shoulder damage, weight of pieces and unsound kernel were assessed for raw kernels, and colour, mottled colour and surface damage for roasted kernels. RESULTS: Shoulder damage, weight of pieces and unsound kernel for raw kernels increased significantly between the dehusker and the final silo. Roasted kernels displayed a significant increase in dark colour, mottled colour and surface damage during on‐farm handling. CONCLUSION: Significant loss of macadamia kernel quality occurred on a commercial farm during sorting and storage of nuts‐in‐shell before nuts were consigned to a processor. Nuts‐in‐shell should be dried as quickly as possible and on‐farm handling minimised to maintain optimum kernel quality. Copyright © 2010 Society of Chemical Industry     

小麦不完善粒快速检测仪器研究(Online First, Recommended Article)

采用图像分析技术与自动控制技术,将人工智能技术应用于小麦不完善粒检测,研究开发了小麦不完善粒指标的自动快速无损检测仪器。通过验证该仪器检测小麦不完善粒的准确性、重复性、稳定性、台间差等相关性能参数,结果表明：该仪器检测性能稳定,准确性、重复性、稳定性、台间差均符合行业标准要求,操作简单,检测速度快,克服了人工检测主观性强、重复性差、不同人员间检验一致性较差等问题,可实现小麦不完善粒的自动快速无损检测,能够满足粮食收储企业、加工企业和检测机构的检测需要。     

基于图像处理的玉米颗粒霉变程度检测方法研究

提出利用HSV模型快速识别霉变玉米颗粒与霉变等级的方法。基于感染霉菌的玉米颗粒表层会发生颜色褐变、发黑等特点,首先对彩色玉米图像进行图像增强,然后采用V分量的阈值区分正常玉米颗粒与霉变玉米颗粒,再通过H与V分量的阈值区分轻度霉变与严重霉变玉米颗粒。试验表明,该方法对正常玉米颗粒、轻度霉变玉米颗粒、严重霉变玉米颗粒的检测准确率可达93.7%、80%、92.9%以上,能够达到区分检测的目的。     

Discriminating and elimination of damaged soybean seeds based on image characteristics

To identify and eliminate damaged soybean seeds, images of Kaiyu 857 soybean seeds including those with insect damage, mildew, and other defects were acquired with an intelligent camera. After splitting the kernels from the background through using the data fusion, morphological corrosion expansion and a series of image processing algorithms, we extracted eight shape features, three color features and three texture features as the input layer to set up a BP neural network classification model with an average recognition accuracy of 97.25%. The identifying and eliminating device was tested five times with a mixture of 1000 differently damaged soybeans of seeds. The average accuracy rates of identification and elimination for normal, mildewed, insect-damaged, skin-damaged, broken and partly defective kernels reached 99.24%, 98.2%, 96.4%, 85.6%, 92.4% and 85.2% respectively. The efficient processing speed of the device reached 125 grains per minute. The results are of significance for the development of precise selection systems for soybeans or other crop seeds.     

不同规格CTC红碎茶理化与香气成分比较

从感官审评、理化成分、香气成分等方面对4个规格的CTC红碎茶进行分析。结果表明,所有红碎茶干茶色泽以棕褐、茶汤以红亮为主,滋味以浓、强、鲜为突出特点,香气类型为甜香,碎茶3号感官品质评分最高。各类理化成分含量在不同规格红碎茶中有显著差异,水浸出物、茶碱和儿茶素总量随颗粒减小而降低;碎茶3号茶多酚（25.68 mg/g）、咖啡碱（43.67 mg/g）和儿茶素总量（96.93 mg/g）含量最高,碎茶5号黄酮总量（20.30 mg/g）最高。GC-MS从4个规格红碎茶中检测到醇类（44.11%~51.84%）、酯类（15.31%~18.96%）、醛类（5.99%~11.56%）等9类共100种香气物质,共有香气成分21种;碎茶3号香气成分与其他规格差异最大,香气物质种类最为丰富;碎茶2号和末茶香气成分最接近,碎茶5号香气物质种类最少。通过分析,不同规格CTC红碎茶在感官审评、化学成分和香气成分方面各有差别,初步探明不同规格CTC红碎茶的品质差异。     

Influence of harvest date and geographical location on kernel symptoms, fungal infestation and embryo viability of malting barley

This study focused on the influence of harvest timing and geographical location on spring malting barley (potential malting variety) in terms of fungal infestation and seedling viability in symptomatic and asymptomatic kernels. The study was conducted over two consecutive crop years (2003-2004) at four locations in Slovakia. The following categories of the barley kernels were used for the assessment: kernels with black point symptoms (BPK), kernels with black cover on their surface (BCK), pink-colored kernels (PCK) and asymptomatic kernels (AK). The occurrence of various fungi was detected in all kernel categories ranging from 60 to 100%. The most frequent fungal contaminants were members of the genus Alternaria, recovered from BPK in the range of 52.6-69.85%, BCK (62.36-73.28%) and AK (55.35-69.58%). The prevalence of Alternaria spp. was recorded for each harvest time. Other dematiaceous fungi, Epicoccum nigrum and Cochliobolus sativus were found with medium frequency in the same three categories. However, C. sativus was recovered with a higher frequency in BPK (2.6-25.3%). In PCK, the most commonly recovered fungi were species of Fusarium with F. avenaceum (59.2-93.2%) as the most prevalent. The infestation of kernels by fungi from other genera showed only low frequency within all the kernel categories investigated. There was no significant influence of the year and location on the proportion of the symptomatic kernel categories. Kernel germination was inhibited to a greater degree in the wetter and colder year. The strongest inhibition of germination ranging from 2.35 to 22.45% was recorded in PCK in all locations and both years. Germination declined from PCK to BCK (43.2-90.32%), and BPK (45.6-91.61%), while it was highest in AK at all harvest times (63-93.6%). This study found that the black covering symptoms (BCK) caused greater damage to viability of the kernels than black point symptoms (BPK). The delayed harvest time resulted not only in increasing numbers of discolored kernels and undesirable symptomatic kernel fractions but also in lower germination of the kernels tested.     

High-speed bichromatic inspection of wheat kernels for mold and color class using high-power pulsed LEDs

High-speed optical sorting of seeds in commercial processing is routinely practiced for removal of discolored seeds, seeds from volunteer plants, and non-seed objects. Sorters are conventionally based on monochromatic or bichromatic light from broad wavebands in the visible and near-infrared regions of energy. A particular challenge for these devices has been the recognition and removal of wheat kernels that have been damaged by the mold caused by the fungal disease Fusarium Head Blight. Previous research using an off-the-shelf bichromatic design on Fusarium-damaged wheat kernels demonstrated that approximately half of damaged kernels were positively detected. The research described herein examines an alternative design for bichromatic lighting and applies this design to two scenarios: sound vs. Fusarium-damaged wheat and red vs. white wheat. The new design utilizes two high-power (HP) LEDs and one silicon photo diode detector. The LEDs are flashed in alternating sequence at high frequency (2,000 Hz), such that during the half-cycle time period (0.25 ms) that each LED is on, reflected energy readings at a 10× sampling frequency are captured from a kernel in flight. This permits the capture of approximately 20 cycles of pulsed light during the time the free-falling kernel passes through the field of view of a fiber optic probe. A linear discriminant analysis (LDA) classification algorithm was applied that used two values derived from the reflected energy readings. Based on the new design, the accuracy of sound vs. Fusarium-damaged classification was 78% on average; for red vs. white wheat classification, the average accuracy was 76%. Although these accuracy values are not at the level as that obtained from LDA models that utilize reflected energy readings at two wavelengths from stationary kernels (95% and 92% for sound vs. Fusarium-damaged and red vs. white, respectively), the new design offers an improvement over conventional bichromatic designs. Mention of trade names or commercial products is solely for the purpose of providing specific information and does not imply endorsement or recommendation by the USDA.     

Several physical properties of aflatoxin-contaminated pistachio nuts: Application of BGY fluorescence for separation of aflatoxin-contaminated nuts

The primary objective was to evaluate and find a proper method for visual identification of aflatoxin-contaminated pistachio nuts. The feasibility of using bright greenish yellow fluorescence (BGYF) in pistachio nut as a discriminating factor for identification of Aspergillus flavus-infested nuts, at harvest and in post-harvest, is investigated. Results show a strong relationship between BGYF and aflatoxin content at harvest. The factors affecting the application of this method in post-harvest stages are also discussed. The relationship between inside-brown kernels and aflatoxin presence is confirmed. At harvest, the brown kernels are a subdivision of fluorescent fraction. The share of different pistachios based on hull types (with sound hull, growth split and early-split) in contamination is studied. The early-split nuts are the most contaminated nuts, growth split nuts are less contaminated, and pistachios with sound hulls are almost clean. The effect of inappropriate handling on the percentage of fluorescent nuts is studied. The percentage of visible mould in samples is observed which shows a good relationship with the presence of BGY fluorescence.     

小麦图像检测技术研究进展

图像检测技术具有检测结果客观、分析速度快、不破坏样品等优点,在农产品检测等多种领域越来越受到人们的关注。通过对图像检测技术在小麦质量与品质检测等方面的应用进行综述,介绍了该检测技术在小麦品种分类、小麦与类似谷物的识别、小麦品质检测与分级、小麦不完善粒的检测、小麦角质和硬度的测定等方面的研究进展,并展望了图像检测技术在小麦综合品质判定方面的发展前景,以期为图像检测技术在小麦品质与质量检测方面更广阔的应用提供参考。     

An update on the processing of high‐protein rice products

The component of protein in rice, at 7–9% by weight, is relatively low, but the total amount of rice protein potentially available is significant because the production of rice worldwide, at 380 million tons annually, is huge. Rice proteins are recognized as nutritional, hypoallergenic, and healthy for human consumption, and rice protein products have been in demand in recent years. However, because of difficulties in the processing, rice protein products, particularly high‐protein content ones, have not been readily available. Two of the main sources of rice protein, rice bran and, to a lesser extent, broken rice kernels, have been under‐used and under‐priced. This report provides an update on the processing of these sources for rice proteins. Methods of protein processing are highlighted including the traditional alkaline extraction, enzyme‐assisted extraction, and the novel uses of physical treatment prior to water extraction. Also discussed are effects of processing on the functional and nutritional properties of rice protein.