梨和苹果糖度在线检测通用数学模型研究

采用可见/近红外光谱技术在线检测水果糖度,每个水果品种要单独建模,模型升级维护耗时费力。探讨建立苹果、梨等薄皮水果可溶性固形物(SSC)在线检测通用数学模型的可行性。利用自行设计的可见/近红外漫透射光谱在线检测系统,在积分时间80 ms、单线速度5个/s的条件下,采集新梨7号、砀山酥梨、玉露香梨和富士苹果四种水果的可见/近红外漫透射光谱。分析了四种水果的可见/近红外漫透射光谱响应特性,采用变异系数法和连续投影算法,筛选通用数学模型建模用光谱变量,并建立了偏最小二乘和最小二乘支持向量机梨与苹果梨通用数学模型。采用新样品评价模型的预测能力,变异系数法筛选光谱波段建立的偏最小二乘通用数学模型预测精度最高,通用模型预测梨和苹果梨模型预测均方根误差分别为0.49%和0.55%,通用模型预测相关系数分别为0.88和0.93;独立模型预测新梨7号、玉露香梨、砀山酥梨和富士苹果的预测相关系数分别为0.93,0.91,0.88和0.95,预测均方根误差分别为0.40%,0.42%,0.41%和0.46%。通用数学模型的预测精度略低于每个品种的独立数学模型,但是通用模型的通用性高于单一模型。实验结果说明采用变异系数法结合偏最小二乘法建立薄皮水果在线检测通用数学模型,实现四种水果糖度在线检测是可行的。     

基于Slope/Bias算法的相近种类水果模型传递研究

为了提升便携式近红外仪器中单一水果模型应用的广泛性,创新性的将不同仪器间模型传递的思想应用在不同种类水果间可溶性固形物(soluble solid content,SSC)的模型传递。基于苹果、梨、桃三种水果在SSC含量范围、果型大小以及果皮厚度等的相近物理化学特性,提出利用简单的斜率/截距(Slope/Bias)算法对苹果SSC的偏最小二乘(partial least square,PLS)模型进行传递,仅用少量的梨和桃样品即可将苹果SSC模型应用于其SSC值的预测,更快捷方便且节约成本。对于梨样品,用35个标准样品,预测集均方根误差(root mean square error of prediction,RMSEP)值由直接预测的1.009°Brix降为0.565°Brix;对于桃样品,用40个标准样品,RMSEP由直接预测的1.726°Brix降为0.677°Brix。为了验证该模型传递方法的可行性,通过斜率/截距算法,采用梨和桃模型对其他两种水果的SSC进行预测,其中利用建立的梨SSC模型,经斜率/截距算法模型传递后,对于苹果样品,用30个标准样品,RMSEP值达到0.597°Brix,对于桃样品,用40个标准样品,RMSEP值达到0.689°Brix;利用建立的桃SSC模型,经斜率/截距算法模型传递后,对于苹果样品,用35个标准样品,RMSEP值达到0.654°Brix,对于梨样品,用30个标准样品,RMSEP值达到0.439°Brix。研究结果表明:斜率/截距(Slope/Bias)方法可用于苹果、梨、桃等相近种类水果间的模型传递,为近红外光谱仪在相似种类物质间的预测提供了新思路。     

Point-wise and whole-field laser speckle intensity fluctuation measurements applied to botanical specimens

Based on multi-scattering speckle theory, the speckle fields generated by plant specimens irradiated by laser light have been studied using a pointwise method. In addition, a whole-field method has been developed with which entire botanical specimens may be studied. Results are reported from measurements made on tomato and apple fruits, orange peel, leaves of tobacco seedlings, leaves of shihu seedlings (a Chinese medicinal herb), soy-bean sprouts, and leaves from an unidentified trailing houseplant. Although differences where observed in the temporal fluctuations of speckles that could be ascribed to differences in age and vitality, the growing tip of the bean sprout and the shihu seedling both generated virtually stationary speckles such as were observed from boiled orange peel and from localised heat-damaged regions on apple fruit. Our results suggest that both the identity of the botanical specimen and the site at which measurements are taken are likely to critically affect the observation or otherwise of temporal fluctuations of laser speckles.     

Visualization and Quantification of Vascular Structure of Fruit Using Magnetic Resonance Microimaging

Three-dimensional magnetic resonance (MR) images of four kinds of fruit (Japanese apricot, peach, Japanese pear, and apple) were acquired throughout their growing periods (from April to August) to visualize and quantify their vascular structures. The vascular structures were emphasized using T ₁-weighted spin-echo or gradient-echo pulse sequences, visualized using a maximum-intensity projection technique, and quantified using an image-segmentation program. As a result, we demonstrated that the vascular volume changed almost in proportion to the whole volume and that larger fruits generally had larger proportionality constants. We therefore concluded that MR microimaging is a powerful tool for studies of the vascular structure of fruit.     

Nondestructive testing of external defects on Nanguo pear

In view of the common pests and diseases and irregularly shaped fruits of Nanguo pear, this paper fused the spectral information and image features to realize the rapid nondestructive testing and recognition of the external defects on Nanguo pear by hyperspectral imaging technology. Backpropagation neural network and support vector machine model was established to identify external defects, which are commonly used in classification and pattern recognition. The testing results show that recognition effect of support vector machine is better than backpropagation neural network. Among them, the recognition accuracy of fruits damaged by insects and rotten fruits of Nanguo pears reaches 100%. This study provides a theoretical basis for developing online grading system and quality detection of Nanguo pear based on multispectral imaging technique.     

不同红梨品种果实中营养元素含量的光谱测定

用火焰原子吸收光谱法对美人酥、红酥脆和满天红3个红梨品种果实中K,Fe,Zn,Ca,Cu,Mg和Mn七种营养元素含量进行了测定分析。各元素在实验范围内,线性关系良好,回收率在98%～104%之间,结果较为满意。3个红梨品种果实中均含有丰富的营养元素,且含量均是K＞Fe＞Zn＞Ca＞Cu＞Mg＞Mn。不同品种果实中,K,Zn,Mg和Mn含量为美人酥＞红酥脆＞满天红;Fe和Cu含量为红酥脆＞满天红＞美人酥;Ca含量为满天红＞红酥脆＞美人酥;七种营养元素总量为红酥脆＞美人酥＞满天红。     

随机森林算法的水果糖分近红外光谱测量

近年来,有关水果糖分等内部品质的近红外光谱测量方法研究很多,并有部分商业化仪器问世。但由于近红外光谱复杂多变,模型的传递性较差,往往所建模型只能针对特定品种甚至特定产地的水果。随机森林(RF)是一种基于决策树的集成算法,通过对分类回归树(CART)模型的集成来提高预测精度。相对于偏最小二乘法(PLS),多元线型回归法(MLR)等方法,随机森林回归方法对非线性数据的解析能力较强。考虑到RF模型的随机性, 通过调试决策树数量(ntree)和分裂变量数目(mtry)等变量来进行模型优选。尝试使用随机森林对不同种类的水果(苹果、梨)糖分进行预测。实验表明,对于同一种类的水果,随机森林和PLS的建模和预测结果均较好。但对于不同种类的水果,随机森林明显增加了模型的预测能力,将建模R2由PLS的0.878提高到了0.999,将建模的RMSEC由0.453降低到了0.015。经过独立的预测集样品对最优RF模型进行检验,预测R2由PLS的0.731提高到为0.888,预测RMSEP由1.148降低到0.334。随机森林在对多种水果糖分预测时,具有明显的优势。这一研究证明了随机森林有望应用于多种水果糖分的近红外光谱测定,进而解决模型的普适性和传递性问题。     

Decomposition of biospeckle images in temporary spectral bands

We present a method of analysis of images of dynamic speckle based on the filtering in frequency bands of the temporary history of each pixel. Butterworth filters are applied to the temporary evolution, and different images are constructed showing the energy in each frequency band. Different degrees of activity of the sample in study, presumably attributed to different origins, are found. The method is exemplified with images of bruising damage in fruits and of biological activity in germinating corn seeds. It is found that the activity in the bruised region of an apple differs from the activity of healthy regions in a certain characteristic frequency range. The activity of the embryo can also be distinguished from that of the endosperm in corn seeds during germination.     

高光谱成像技术的库尔勒梨早期损伤可视化检测研究

利用高光谱成像技术对库尔勒梨早期损伤进行快速识别检测。以60个库尔勒梨为研究对象,采集380～1 030 nm波段范围内完好样本和损伤后1～7天样本的480幅高光谱图像。提取图像中感兴趣区域(ROI)的平均光谱信息,利用小波变换(WT)对光谱数据进行去噪平滑,将去噪后的全部样本按2∶1的比例分成建模集(320个)和预测集(160个)。利用二阶导数从全谱信息中提取出19个特征波长,分别基于全谱和提取出的特征波长对建模集和预测集进行支持向量机(SVM)建模分析。结果表明,基于全谱和特征波长的判别分析模型中,两者预测集的识别率都达到93.75%,表明提取的特征波长包含了光谱数据中的关键信息。然后,基于特征波长运用波段比运算挑选最佳波段比,根据波段比F值的分布确定光谱图像分割的最佳波长684和798 nm。对最佳波段比(684/798 nm)下的图像,利用选择性搜索(SS)对高光谱图像中样本的完好和损伤区域进行分割,从分割结果来看,1～7天损伤样本的受损区域能够被准确检测出来。研究结果表明：基于高光谱成像技术对库尔勒梨进行损伤鉴别是可行的,该研究所获得的特征波长和波段比为研发在线实时的库尔勒梨损伤检测系统提供支撑。     

基于近红外光谱分析技术测定库尔勒香梨硬度

采用近红外(NIR)漫反射光谱法对新疆特色梨果库尔勒香梨的五种不同果（包括青头、粗皮、脱萼、宿萼、突顶果）的硬度进行测定。由于近红外光谱数据量大且原始光谱噪声明显、测定水果时散射严重等导致光谱建模时关键波长变量提取困难。以新疆库尔勒香梨为研究对象,为了有效地消除固体表面散射以及光程变化对NIR漫反射光谱的影响,首先采用标准正态变量变换（SNV）和多元散射校正（MSC）对库尔勒香梨的原始光谱进行预处理。为寻找适合近红外光谱检测库尔勒香梨硬度的最佳特征波长筛选方法,进行香梨近红外光谱的特征波长变量选择方法的比较与研究。研究比较了两种特征波长筛选方法对库尔勒香梨硬度偏最小二乘法（PLS）建模精度的影响。同时使用反向偏最小二乘（BiPLS）和遗传算法结合反向偏最小二乘（BiPLS-GA）在全光谱范围内筛选香梨硬度的特征波长变量,将校正均方根误差（RESMC）、预测均方根误差（RESMP）以及决定系数（R2）作为模型的评价标准,并最终确定最优波段选择方法及最佳预测模型。基于选择的特征波长变量建立的PLS模型（BiPLS-GA）与全光谱变量建立的PLS模型进行比较发现BiPLS-GA模型仅仅使用原始变量中6.6%的信息就获得了比全变量PLS模型更好的库尔勒香梨硬度的预测结果,其中R2,RMSEC和RMSEP分别为0.91,1.03和1.01。进一步与基于反向偏最小二乘算法（BiPLS）获得的特征变量建立的PLS模型比较发现,BiPLS-GA不仅可以去除原始光谱数据中的无信息变量,同时也能够对共线性的变量进行压缩去除,使得建模变量从301个减少到20个。极大地简化模型的同时有效地提高了模型的预测精准度和稳定性。因此该方法能够有效地用于近红外光谱数据变量的选择。证明了近红外光谱分析技术结合BiPLS-GA模型能够高效地选择出建模变量,去除与库尔勒香梨硬度无关的近红外光谱信息,显著地提高库尔勒香梨硬度定量模型的预测精度。这不仅为新疆地区特色梨果库尔勒香梨的快速、精确、无损优选分级提供一定的技术支持,同时也为基于近红外光谱分析技术预测水果内部品质的研究提供了参考。     

番茄碰伤和可溶性固形物近红外光谱同时在线检测

可溶性固形物和碰伤是影响番茄品质的两个主要因素。研究的目的是探索可见近红外漫透射光谱同时在线检测番茄碰伤和可溶性固形物的可行性。在单通道送果速度5个每秒条件下,采集番茄近红外漫透射光谱。对比分析碰伤与正常番茄样品的近红外漫透射光谱特性,结果表明,碰伤与正常番茄样品的近红外漫透射光谱在光强上存在明显差异,碰伤果光强要强于正常果,其原因可能是碰伤后果肉变软,透光性变强;在650和675 nm处碰伤果比正常果要多两个吸收峰,可能是碰伤后,番茄表皮颜色发生变化所致。选取贡献率占比最多的前三个主成数,对正常果与碰伤果近红外漫透射光谱主成分定性分析,正常果与碰伤果不能有效聚类,故近红外漫透射光谱主成分定性分析效果不明显,需选择建立高维近红外漫透射光谱定性判别模型。故建立了碰伤番茄样品的近红外漫透射光谱偏最小二乘定性判别模型,误判率为0%,能正确判别碰伤果,故选用碰伤番茄样品的近红外漫透射光谱偏最小二乘定性判别模型作为番茄碰伤果在线剔除分选模型。通过对未参与建模的样品进行验证,能正确识别出碰伤果。经近红外漫透射光谱偏最小二乘定性判别模型剔除碰伤果后,按照可溶性固形物指标进行分级。分别使用全部波段和606～850 nm的波段进行建模预处理,且对全部波段和606～850 nm波段光谱进行2阶导数预处理,前后平滑设为9,利用连续投影算法与遗传算法优选可溶性固形物的光谱建模变量,对比发现,利用未经算法筛选过的606～850 nm波段光谱变量进行建模,效果最好,建立了可溶性固形物在线检测模型,预测集均方根误差为0.43 Brix°。采用未参与建模的样品进行碰伤和可溶性固形物同时在线检测验证,碰伤样品的分选准确率达96%,可溶性固形物样品的分选准确率达91%。表明：番茄碰伤和可溶性固形物近红外漫透射光谱同时在线检测是可行的。     

方位和品质对南疆果品空间特性光谱影响及反演方法研究

高光谱无损检测技术在果品定量无损检测中应用广泛,以冬枣、红提、香梨三种果品空间特性光谱为研究目标,探索空间特性光谱的影响因素和反演方法,为提高户外果品无损检测精度提供了一种新思路。分别提取三种果品的光谱库并计算空间特性光谱,依次使用马氏距离、浓度残差等预处理方法以及竞争性自适应权重取样算法选取特征波长,将处理后的三种果品空间特性光谱分别与品质（糖度、水分）和方位（探测角、方位角、相位角）建模,建模结果如下：三种果品（按照冬枣、红提、香梨的顺序）与糖分模型的相关系数r分别为：0.853 3,0.822 7和0.913 3;水分模型的相关系数r分别为：0.741 3,0.784 7和0.891 3;探测角模型相关系数r分别为：0.985 6,0.992 7和0.974 7;方位角模型相关系数r分别为：0.941 8,0.910 5和0.936 9;相位角模型相关系数r分别为：0.960 9,0.957 0和0.956 3。可以看出,不同果品方位模型相关性都明显高于品质模型相关性,因此方位因素是影响空间特性光谱的主要原因。使用Roujean模型和Walthall模型分别对不同方位的空间特性光谱进行反演,反演结果如下：使用Roujean模型反演三种果品（按照冬枣、红提、香梨的顺序）空间特性光谱时R2分别为0.934 4,0.928 1和0.830 6;r分别为0.990 2,0.983 9和0.969 1;RMSEP分别为0.030 9,0.048 7和0.062 7;平均模型误差分别为7.27%,11.02%和8.61%。使用Walthall模型描述不同果品空间特性光谱时R2分别为0.943 3,0.859 7和0.839 0;r分别为0.991 8,0.971 8和0.970 2;RMSEP分别为0.036 6,0.066 1和0.068 7;平均模型误差分别为6.19%,15.40%和7.84%。可以看出,Roujean模型可以很好的描述冬枣和红提的空间特性光谱,也可以较好的描述香梨空间特性光谱;Walthall模型可以很好的描述冬枣空间特性光谱,也可以较好的描述红提和香梨空间特性光谱。综上所述,在今后试验中可以使用Roujean模型反演红提和香梨的空间特性光谱,使用Walthall模型反演冬枣的空间特性光谱,进而提高户外果品户外果品无损检测精度。     

超微弱发光用于番茄果实冷害发生程度的无损监测

超微弱发光测定法可快速、灵敏地反映生物体在逆境下的生理状态,已开始得到广泛应用,但有关其在采后园艺产品冷害发生监测中的应用尚未见报道.文章通过分析番茄果实在不同冷害状态下的超微弱发光规律,及其与冷害程度问的关系,探讨超微弱发光测定法在监测采后番茄果实冷害中应用的可能性.结果表明,果实超微弱发光强度随冷害程度的提高而上升,且在表现冷害症状前,冷害果实的超微弱发光值已显著高于正常果实(P<0.01).不同冷害程度下果实的冷害发病率、冷害指数分别与超微弱发光值的相关系数达到0.901 6和0.957 7(P<0.01),证明超微弱发光可以准确反映番茄果实冷害程度.     

Comparative Investigations on the Fate of Dinobuton and Dinobutonmethyl in Different Plants

It was investigated the phytotoxicity as well as the fate of dinobuton and dinobutonmethyl in bean and cucumber plants and in apple fruits. Dinobutonmethyl is more phytotoxic than dinobuton. Generally dinobuton and dinobutonmethyl behave similarly after application on the apple surface and also after treatment of the leaves of bean and cucumber plants. After 3–5 days only about 50% of the applied radioactivity were found on the surface of the systems. Probably dinobuton and dinobutonmethyl as well as dinoseb as hydrolysis product are able to leave the plant systems unchanged.     

Influence of postharvest ultrasounds treatments on tomato (Solanum lycopersicum,cv. Zinac) quality and microbial load during storage

Whole tomato fruits were treated at ultrasonic power levels from 10% to 100%, and at a constant frequency of 45 kHz, for different times (1–19 min). A central composite rotatable design (CCRD) was applied to optimise ultrasonic treatments for tomato quality (colour, texture and total phenolic content (TPC)) maintenance. According to response surface analysis, the optimal treatment parameters were 55%_10 min, 80%_15 min and 100%_19 min. At these conditions, and especially at higher power levels, a maximum retention of colour and texture, as well as an increase of TPC and microbial reduction were obtained in comparison with untreated fruits during 15 storage days at 10 °C. The ultrasounds treatment was found to be effective in delaying colour development and texture losses, preserving sensorial quality of whole tomato, with increase of TPC and microbial load reduction. Moreover, this postharvest treatment can be used as an alternative for extending fresh fruits shelf-life.     

Effect of plasma-activated water and buffer solution combined with ultrasound on fungicide degradation and quality of cherry tomato during storage

The purpose of this study was to examine plasma-activated buffer solution (PABS) and plasma-activated water (PAW) combined with ultrasonication (U) treatment on the reduction of chlorothalonil fungicide and the quality of tomato fruits during storage. To obtain PAW and PABS, an atmospheric air plasma jet was used to treat buffer solution and deionized water at different treatment times (5 and 10 min). For combined treatments, fruits were submerged in PAW and PABS, then sonicated for 15 min, and individual treatment without sonication. As per the results, the maximum chlorothalonil reduction of 89.29% was detected in PAW-U₁₀, followed by 85.43% in PABS. At the end of the storage period, the maximum reduction of 97.25% was recorded in PAW-U₁₀, followed by 93.14% in PABS-U₁₀. PAW, PABS, and both combined with ultrasound did not significantly affect the overall tomato fruit quality in the storage period. Our results revealed that PAW combined with sonication had a significant impact on post-harvest agrochemical degradation and retention of tomato quality than PABS. Conclusively, the integrated hurdle technologies effectively reduce agrochemical residues, which helps to lower health hazards and foodborne illnesses.     

纳米增强激光诱导击穿光谱的苹果表面农药残留检测

水果表面的农药残留严重危害身体健康,而常规检测方法需要采样处理,耗时、费力。激光诱导击穿光谱技术具有多元素分析和原位测量的能力,在水果表面农药残留检测方面潜力巨大。但是较差的检测灵敏度,限制了此技术在水果表面微量有害元素检测方面的应用。提高激光诱导击穿光谱系统的检测能力是目前的研究热点领域,研究了纳米粒子表面增强技术对苹果表面残留的毒死蜱农药的激光诱导击穿光谱信号的增强效果。通过在被测样品表面涂抹币族金属纳米粒子,然后利用激光诱导击穿光谱激发样品表面,对诱导出的原子发射光谱信号进行测量,实验结果表明,苹果表面涂抹金属纳米粒子后毒死蜱农药中磷元素的特征峰有4倍的增强。此方法的应用对提高果蔬表面微量有害元素的检测能力具有重要意义。然后我们对币族金属纳米粒子的增强效果进行了优化。研究了金纳米粒子和银纳米粒子的增强能力,同时对纳米粒子的粒径的增强效果进行了比较,通过对20 nm的金纳米粒子、20 nm的银纳米粒子和80 nm的银纳米粒子的增强效果比较,发现80 nm的银纳米粒子对苹果表面毒死蜱农药光谱的增强效果最好。研究了信号采集延时时间对光谱信噪比的影响,确定了0.2 μs的延时时间可以获得较为理想的信噪比。在以上研究的基础上,采用最优的实验参数(80 nm银粒子增强、0.2 μs的延时时间),以毒死蜱中磷元素在213.62,214.91,253.56和255.33 nm处的特征峰峰强作为依据,对苹果表面残留的毒死蜱农药进行了定量化分析。分别采集了毒死蜱残留浓度分别为30,20,15,12,10和6 μg·cm-2的苹果表面的LIBS光谱,然后分别利用磷元素的四个特征峰峰强进行了定量化曲线拟合,结果发现LIBS对残留的毒死蜱具有很好的定量化预测能力,R2在0.89以上。根据定量化拟合曲线,探讨了纳米增强LIBS的检测限,计算得到,利用纳米增强LIBS技术测量苹果表面的毒死蜱农药最低可以做到1.61 μg·cm-2的检测限。研究证明了金属纳米粒显著提高了LIBS对苹果表面农药残留的检测灵敏度。     

微波消解/ICP-MS法检测八种梨果实中主要矿质元素含量

采用ICP-MS法测定了八种梨果实中的钠、钾、钙、镁、锌、铜、铁、锰、硼等元素的含量。不同品种梨果实中主要元素组成基本相同, 但各种元素的含量存在一定的差异。其中,钾的含量最高, 其次是镁、钙、钠等元素。采用欧盟果蔬汁工业协会(AIJN)提供的果汁质量评价参考范围作参照标准,梨果实中钾、镁和钙元素含量高于该标准,钠元素含量低于该标准,其余元素含量均接近于该标准的规定。     

Rapid Detection of Infestation of Apple Fruits by the Peach Fruit Moth, Carposina sasakii Matsumura, Larvae Using a 0.2-T Dedicated Magnetic Resonance Imaging Apparatus

Infestation of harvested apple fruits by the peach fruit moth (Carposina sasakii Matsumura) was studied using a dedicated magnetic resonance imaging (MRI) apparatus equipped with a 0.2-T permanent magnet. Infested holes on the three-dimensional (3-D) images tracked ecological movements of peach fruit moth larvae within the food fruits, and thus in their natural habitat. Sensitive short solenoid coil and surface coil detectors were devised to shorten measurement times. The short solenoid coil detected infestation holes at a rate of 6.4 s per image by the single-slice 2-D measurement. The multi-slice 2-D measurement provided six slice images of a fruit within 2 min taken by the two detectors. These results indicate that the 0.2-T MRI apparatus allows one to distinguish sound fruits from infested ones, and also as a means for plant protection and the preservation of natural ecological systems in foreign trade.     

Fabrication of Au nanorod‐coated Fe3O4 microspheres as SERS substrate for pesticide analysis by near‐infrared excitation

Au nanorods coated Fe3O4 (Fe₃O₄@NRs) microspheres were designed as functional surface‐enhanced Raman scattering substrate with a feature of magnetic property and used for detection of pesticide residues that are annually used in agriculture by near‐infrared (NIR) excitation. With this strategy, the Fe₃O₄ microspheres were synthesized by hydrothermal method and surface functionalized with polyethylenimine, and then coated with Au nanorods densely. The Raman spectra were carried out by NIR excitation and 4‐ATP was chosen as the probe molecule. The results showed a good SERS activity of the Fe₃O₄@NRs microspheres. Moreover, this substrate could be used for pesticide analysis by portable Raman spectrometer with NIR excitation. Especially, the microspheres could be transferred from pesticides contaminated fruits peel to specially cleaned glass slide with the aid of the external magnetic field, by which the strong fluorescence of the apple components can be avoided while performing the pesticide analysis of fruits peel. Copyright © 2015 John Wiley & Sons, Ltd.