马铃薯黑心病和单薯质量的透射高光谱检测方法

针对单一检测技术不能同时检测马铃薯内外品质的多项指标,采用透射高光谱成像技术并融合光谱和图像信息,对其内部黑心病、质量指标进行检测。通过透射高光谱成像系统获取266个样本高光谱图像（400～1000 nm）,并提取光谱和图像二者信息。采用不同变量选择方法对光谱进行变量选择,用9个光谱变量建立检测马铃薯黑心病偏最小二乘判别分析（partial least squares discriminant analysis, PLS-DA）模型与质量偏最小二乘回归（partial least squares, PLS）模型;提取样本透射高光谱图像的面积信息,建立基于光谱-图像的检测马铃薯质量PLS模型。试验结果表明,黑心样本识别率为100%,识别最小黑心面积为1.88 cm2;基于光谱-图像所建立质量检测模型预测效果较好,其预测集相关系数(Rp)为0.99,预测均方根误差(RMSEP）为10.88。结果表明：采用透射高光谱成像技术并融合图像和光谱信息对马铃薯内部黑心病、质量同时进行检测是可行的。     

基于局部加权回归的土壤全氮含量可见-近红外光谱反演

全氮是土壤肥力的重要指标,对作物产量具有决定性作用,采用土壤可见-近红外(Vis-NIR)光谱预测技术及时获取土壤全氮含量信息具有重要意义。采用来自5省的450个土壤样本来验证局部加权回归方法(LWR)结合Vis-NIR光谱技术预测大面积土壤全氮含量的适用性。结果表明,LWR模型的预测效果优于偏最小二乘回归(PLSR)、人工神经网络(ANN)和支持向量机(SVM),选取主成分数为5,相似样本为40时,模型验证的决定系数(RP2)为0.83,均方根误差(RMSEP)为0.25 g kg-1,测定值标准偏差与标准预测误差的比值(RPD)达到2.41。LWR从建模集中选取与验证样本相似的土样作为局部建模样本,降低了差别大的样本对模型的干扰,从而提高了模型的预测能力。因此,LWR建模方法通过大范围、大样本土壤光谱数据进行大尺度区域的全氮等土壤属性预测时能够发挥更好的作用。     

大白菜中马拉硫磷农药的表面增强拉曼光谱快速检测

为了检测大白菜中马拉硫磷农药残留,该文采用表面增强拉曼光谱技术结合化学计量学方法建立马拉硫磷残留的快速检测模型。采用硫酸镁、N-丙基乙二胺、石墨化炭黑和C18去除大白菜中蛋白质、脂肪、碳水化合物等物质的影响。利用不同预处理方法对原始光谱信号进行预处理,建立大白菜中马拉硫磷残留的偏最小二乘模型。研究发现,大白菜中马拉硫磷的检测浓度达到1.082 mg/L以下;归一化预处理后建立的模型预测性能最好。配制5个未知浓度样本验证模型的准确度,预测值与真实值相对误差的绝对值为0.70%~9.84%,预测回收率为99.30%~109.84%;配对t检验的结果表明样本的预测值与真实值之间无明显差异,说明模型是准确可靠的。结果表明,SERS(surface-enhanced Raman spectroscopy)方法可以实现大白菜中马拉硫磷残留的快速检测。     

傅立叶变换近红外光谱法检测白酒总酸和总酯

该文研究白酒总酸和总酯的快速检测技术,通过解析不同白酒样品的近红外光谱图,对光谱数据进行不同的处理,结果表明：用一阶导数预处理光谱,谱区选择6102～5446 cm^-1,利用人工神经网络与傅立叶变换近红外光谱相结合,采用内部交叉验证法建立模型,效果较好。其中,总酸模型的决定系数为96.73%,内部交叉验证均方根差为0.048 g/L;总酯模型的决定系数为99.58%,内部交叉验证均方根差为0.085 g/L;进一步对总酸和总酯的模型进行验证和评价,结果表明总酸模型验证集的相关系数为99.2%,预测标准偏差为0.074 g/L;总酯模型验证集的相关为99.7%,预测标准偏差为0.134 g/L,表明建立的模型可靠,预测效果好,能满足白酒生产中总酸和总酯的快速检测要求。     

土壤盐渍化高光谱特征分析与建模

基于高光谱遥感技术快速、无损的检测优势,以新疆渭干河-库车河三角洲绿洲为例,探讨利用反射光谱来预测土壤含盐量的可行性。利用野外采集的土壤样本,在实验室内测得了土壤含盐量及原始光谱反射率。利用光谱分析技术计算高光谱指数,与土壤样本含盐量进行相关性分析,筛选出土壤含盐量的光谱特征波段,基于逐步多元线性回归和偏最小二乘回归建立土壤盐分动态监测模型。通过精度检验,结果表明:基于偏最小二乘回归方法,以对数二阶微分光谱特征波段所构建的盐渍化遥感监测模型最优,模型的稳定性和预测精度最高。利用反射光谱来预测土壤含盐量可实现区域尺度上的土壤盐渍化实时监测和评价。     

利用高光谱透射图像检测鸡种蛋早期孵化

为了检测鸡种蛋孵化前期胚胎发育情况,构建了高光谱图像采集系统,在400～1000nm范围内获取90枚种蛋孵化前3天的高光谱透射图像。通过独立分量方法对高光谱数据进行分析降维,优选出571、614、661、691和716nm共5个特征波长,提取每个波长下的光谱平均值和标准偏差,获得每个样品10个特征变量。为了消除变量之间相关性,利用主成分分析提取了4个主成分变量,在此基础上构建了学习向量量化(LVQ)神经网络判别模型。验证性试验均表明该模型具有较高的稳定度(变异系数为1.7),对第1,2,3天的测试样本判别准确率分别为78.8%,90.3%和98.6%。结果表明高光谱图像技术可以检测种蛋孵化前期胚胎发育情况。     

基于近红外光谱技术的大米品种快速鉴别方法

为探索大米无损检测技术,提出了一种基于可见-近红外光谱技术快速、无损鉴别大米品种的新方法。首先采用主成分分析法对大米品种进行聚类,然后利用小波变换技术提取光谱特征信息,把光谱特征信息作为人工神经网络的输入建立品种识别模型,对大米品种进行鉴别。从每种大米60个样本共计180个样本中随机抽取150个样本（每种50个样本）用来建立神经网络模型,剩下的30个大米样本用于预测。品种识别准确率达到100%。说明所提出的方法具有很好的分类和鉴别作用,为大米的品种鉴别提供了一种新方法。     

基于高光谱图像的茶树LAI与氮含量反演

为了对茶树进行实时、快速、无损的叶面积指数LAI和氮含量检测,该文以英红九号茶树为试验对象,利用便携式高光谱成像仪采集光谱数据、人工破坏性采摘叶片进行叶面积指数的计算以及传统化学方法测量叶片氮含量,比较不同高光谱特征变换形式与LAI和氮含量之间的相关性,并选择其中相关系数较高的高光谱特征变量作为自变量,分别采用线性、指数、对数和抛物线表达式建立LAI和氮含量的回归模型。结果显示:在多种高光谱数据变量建立的模型中,以绿峰反射率R_g为自变量的对数拟合模型最佳,其拟合样本的决定系数R~2和验证样本的均方根误差RMSE值分别为0.9和0.087 6。以植被指数变量VI_4(红边面积/黄边面积)与氮含量建立的指数模型为最佳建模效果,拟合样本的决定系数R~2和验证样本的均方根误差RMSE值分别为0.830 3和0.102 9,研究结果可为茶树叶面积指数LAI和营养成分的无损检测提供参考。     

基于高光谱的北京铁矿区土壤重金属镍元素含量反演研究

运用高光谱数据对北京典型铁矿区土壤重金属镍含量进行建模反演,探索高光谱遥感技术在土壤重金属污染快速监测上应用的可行性。使用便携式地物光谱仪采集研究区土壤样本光谱反射率数据,光谱反射率数据经多种数学变换后,经逐步回归方法筛选最佳特征波段,利用多元线性回归（SLR）和偏最小二乘回归（PLSR）方法建立模型以光谱反射数据对土壤重金属镍元素含量进行反演。基于光谱二阶微分的多元线性回归模型（SD-MLR）的稳定性和精度最高（R2 = 0.842,RMSE = 4.474）,能够良好地预测研究区土壤镍元素含量。光谱数据数学变换能够有效提高其与土壤镍元素含量间的相关性。不同的光谱变换形式建立模型的预测能力和精度有如下关系,光谱二阶微分 > 光谱倒数对数一阶微分 > 光谱一阶微分 > 光谱倒数对数 > 光谱连续统去除 > 原始光谱。采用光谱二阶微分建立多元线性回归模型为研究区土壤镍元素含量反演的最佳模型,可为土壤重金属污染快速监测提供技术参考。     

柿子可溶性固形物含量的可见-近红外光谱检测

为了实现柿子（Diospyros kaki thunb）可溶性固形物含量的快速无损检测,提出了一种采用可见-近红外光谱分析技术无损检测柿子可溶性固形物含量的方法。采用Field Spec 3光谱仪对3种不同品种的柿子进行光谱分析,共获取66个样本数据。利用平均平滑法对样本数据进行预处理,再采用主成分分析法,依据可信度获取光谱的6个主成分数据。将样本随机分成51个建模样本（每种各17个）和15个验证样本（每种各5个）,把6个主成分数据作为BP神经网络的输入变量,柿子的可溶性固形物含量作为输出变量,隐含层的节点数为11,建立3层BP神经网络检测模型,并用该模型对15个验证样本进行预测。结果表明,所建校正模型的校正标准差（SEC）为0.232,对预测集样本可溶性固形物含量的预测相对误差在3%以下,预测值和实测值的决定系数（R2）为0.99,预测标准差（SEP）为0.257。结果表明应用近红外光谱技术结合主成分分析和神经网络算法检测柿子的可溶性固形物含量是可行的。     

Electrooxidation Performance of Aqueous Solution of Nonylphenol Decaethoxylate and Denim Wastewater

The goal of this research was the electrooxidation (EO) of a nonionic surfactant nonylphenol decaethoxylate (NP-10) in aqueous solution and denim wastewater. Three different configuration systems were evaluated in batch cells using a boron-doped diamond (BDD) anode; copper, iron, and BDD were used as cathodes. The EO process was carried out in a batch process, in a glass cell with a capacity of 1000 mL. The anode surface area was 0.0307 m² and 1–3 A of current intensity were applied (3, 6, 10 mA/cm²) with an electrolysis time of 240 min for aqueous solution and 780 min for denim wastewater in order to investigate the degradation of the surfactant. The processes were analyzed in terms of chemical oxygen demand (COD) and total organic carbon (TOC). The maximum mineralization efficiency in aqueous solution for the BDD-Cu electrooxidation system was 92.2% for COD and 45.6% for TOC at pH 2 and 3 mA/cm² of current intensity. For denim wastewater, the removal efficiency was 44.1% for COD and 26.5% for TOC at pH 4.5 and 6 mA/cm² of current intensity, using a BDD-BDD system. The raw and treated (aqueous solution and denim) wastewater were characterized by UV-Vis and infrared spectroscopy.     

Innovative Technology of COD and BOD Reduction from Coffee Processing Wastewater Using Avocado Seed Carbon (ASC)

The study estimated the efficiency of Avocado seed carbon (ASC) for chemical oxygen demand (COD) and biochemical oxygen demand (BOD) reduction from coffee processing wastewater. It was performed under batch mode conditions to investigate the optimum operating conditions and efficiency for COD and BOD reduction with ASC compared with commercial activated carbon (CAC). Adsorption isotherm study was also performed and it was found that the values of regression coefficient (r ²), adsorption capacity (k), and adsorption intensity (1/n) for COD and BOD reduction with ASC were comparable to those of CAC. Under optimum operating conditions, the maximum percentage reduction of COD and BOD concentration using ASC was 98.28% and 99.19%, respectively and with CAC was 99.12% and 99.45%, respectively and hence adsorption capacity of ASC is comparable with that of CAC. Thus, this technique may be a good option for treatment of domestic wastewater.     

Oxygen Transfer and Industrial Wastewater Treatment Efficiency of a Vertically Moving Biofilm System

A vertically moving biofilm system (VMBS) was developed to treat wastewater. In this system, the biofilm grows on a biofilm module consisting of plastic media that is vertically and repeatedly moved up into the air and down into the water. The objectives of this study were to investigate the oxygen transfer efficiency and industrial wastewater treatment performance of the VMBS. The oxygen transfer coefficient (K _L a) depended on the movement frequency (n) of the biofilm module and was proportional to n ^1.67. K _L a values measured were within the range of 0.0001 to 0.0027 s^-1. The VMBS exhibited good carbonaceous removal when treating industrial wastewater produced in a factory manufacturing synthetic fibres. Removal efficiency of filtered chemical oxygen demand (COD) and biological oxygen demand (BOD₅) was up to 93.2 and 97.9%, respectively. The volumetric removal rates of filtered COD and BOD₅ reached 1320 g COD m^-3 day^-1 and 700 g BOD₅ m^-3 day^-1. The areal organic removal rates, based on the surface area of the biofilm substrata, were 16 g BOD₅ m^-2 day^-1 and 39 g COD m^-2 day^-1. No clogging occurred during the experiment. The mean areal biofilm mass increased with increasing the mean areal BOD₅ removal rate. The new biofilm process has such advantages as high carbonaceous oxidation, energy saving, simpleconstruction and easy operation for industrial wastewater treatment.     

牛场废水培养肥壮蹄形藻的工艺参数优化

该研究选用肥壮蹄形藻（Kirchneriella obesa）为试验藻种,以秸秆过滤后牛场废水与BG11培养基的混合物为微藻培养液,在前期单因素试验基础上研究废水添加比、光照强度、光照时间和通气量对微藻干重、氨氮去除率、总磷去除率和Chemical Oxygen Demand (COD)去除率的交互影响,通过Central Composite Design (CCD)中心组合试验得到了相应的影响模型。试验结果表明牛场废水培养肥壮蹄形藻的最优工艺条件为：废水添加比例26%、光照强度9 028 lx、光照时间21.5 h、通气量2.0 L/min。该研究采用优化工艺,在14d的培养周期末期,肥壮蹄形藻的干重达到了1.141g/L,废水中氨氮去除率、总磷去除率和COD去除率分别达到了99.65% 、99.15%、85.83%,响应值指标的预测值和实际值误差均在2%以内。各因素对目标值的交互作用关系为：废水添加比例和光照时间对肥壮蹄形藻干重的交互作用较强,废水添加比例和通气量对肥壮蹄形藻培养液中氨氮去除率、总磷去除率和COD去除率均具有较强的交互作用,而光照强度和光照时间仅对总磷去除率具有较强的交互作用。该研究为牛场废水培养微藻的工业化生产提供了理论参考。     

Aerobic Biological Treatment of Chestnut Processing Wastewater

Chestnut agro-industrial companies consume a high volume of water for washing and processing fruit, generating a large volume of wastewater. This work studied the biodegradation of chestnut processing wastewater through aerobic assays, varying substrate, and biomass concentrations. In general, this wastewater presents a good biodegradability, especially in experiments with relatively low chemical oxygen demand (COD) (0.4 and 0.6?g O₂ L^?1) allowing a COD removal of 85?C90?%. The best results were obtained in the reactor initially loaded with 2?g?L^?1 of biomass and 0.4 or 0.6?g O₂ L^?1 of COD. These experiments also showed high COD removal rates: 4.25 and 3.88?g COD g^?1 volatile suspended solids (VSS) h^?1, respectively. The sedimentation rate, evaluated for different initial values of biomass (1, 2, and 3?g?L^?1), always presented higher values in the experiments with 2 and 3?g?L^?1 of biomass, regardless of the initial COD value used. After comparing different kinetic models (Monod, Contois, and Haldane), it was observed that the Haldane inhibition model satisfactorily describes the COD biodegradation. AQUASIM software allowed calculating the kinetic constant ranges: K _s, 1.59?C6.99?g COD L^?1; ?? _max, 25?C40?g COD g^?1 VSS day^?1; and K _i values, 0.07?C0.11. These kinetic constants corresponds to maximum rates (??*) between 1.48 and 4.25?g COD g^?1 VSS day^?1 for substrate concentrations (S*) from 0.38 to 0.88?g COD L^?1.     

重庆三峡库区农村生活污水排放特征及影响因素分析

为了解重庆三峡库区农村生活污水排放特征,正确评价农村生活污染状况,选取重庆三峡库区3个行政区中不同经济收入水平、有无下水设施的18户典型农户为监测对象,对其生活污水日产生量、pH、COD、TP、TN、TAN等6项指标进行了连续12月的跟踪监测。结果表明,18户对象农户生活污水人均日产生量、pH、COD、TP、TN、TAN等指标的平均水平分别为：15.22 L·cap-1·d-1、6.67、838.93 mg·L-1、4.19 mg·L-1、53.60 mg·L-1、13.39 mg·L-1,具有人均污水日产生量明显低于而各特征指标的浓度明显高于城镇生活污水的特点;COD、TP、TN、TAN浓度随季节的变化表现出夏季浓度较低,冬季浓度较高的规律,且COD、TP、TN、TAN浓度之间具有显著的正相关性。此外,利用SPSS软件对影响农村生活污水排放特性的地域、季节、农户收入水平和有无下水设施等因素进行了分析,其结果显示：农村生活污水人均日产生量的主要影响因素为地域,COD的主要影响因素为地域、收入水平和季节,TP的主要影响因素为地域、有无下水, TN和TAN的主要影响因素为季节。     

Screening of Wheat for Flour Swelling Volume by Near-Infrared Spectroscopy

Selection for starch quality is an important consideration in the breeding of wheat for Asian noodles, particularly Japanese udon, and the flour swelling volume (FSV) test was developed for this purpose. Near-infrared reflectance spectroscopy (NIRS) analysis has also been a key tool in recent years in wheat quality selection. The development and validation of NIRS calibrations for the prediction of FSV on whole grain involved 22 cultivars and breeding lines grown at four locations in two seasons. Eight calibrations were developed, each based on samples from seven trials, with the eighth trial used for validation. Over the eight calibrations, r² between predicted and actual values was 0.56–0.86 (mean 0.74) and the standard error of prediction (SEP) was 0.77–1.65 (mean 1.14) mL/g of dry meal. Separate calibrations were also developed for hard (n = 461), soft (n = 150), and soft+hard grain (n = 616), with standard errors of cross-validation (SECV) of 1.03, 1.39, and 1.21 mL/g of dry meal, respectively. Corresponding r² between predicted and actual values were 0.76, 0.78, and 0.76, respectively. Thus, NIRS offers good potential for the screening of early-generation lines to identify those with high or low FSV.     

基于近红外光谱技术的淡水鱼品种快速鉴别

为探索淡水鱼品种的快速鉴别方法,该文应用近红外光谱分析技术,结合化学计量学方法,对7种淡水鱼品种的判别分类进行了研究。采集了青、草、鲢、鳙、鲤、鲫、鲂等7种淡水鱼,共665个鱼肉样品的近红外光谱数据,经过多元散射校正(multiplicative scatter correction,MSC)、正交信号校正(orthogonal signal correction,OSC)、数据标准化(standardization,S)等20种方法预处理,在1 000~1 799 nm范围内分别采用偏最小二乘法(partial least square,PLS)、主成分分析(principal component analysis,PCA)和BP人工神经网络技术(back propagation artificial neural network,BP-ANN)、偏最小二乘法和BP人工神经网络技术对7种淡水鱼原始光谱数据进行了鉴别分析。结果表明,近红外光谱数据,结合主成分分析和BP人工神经网络技术建立的淡水鱼品种鉴别模型最优,模型的鉴别准确率达96.4%,对未知样本的鉴别准确率达95.5%。模型具有较好的鉴别能力,采用该方法能较为准确、快速地鉴别出淡水鱼的品种。     

Can Vegetal Biopolymers Work as Coagulant–Flocculant Aids in the Treatment of High-Load Cosmetic Industrial Wastewaters?

The aim of this work was to evaluate the role of three biopolymers used as coagulant?Cflocculant aids in the treatment of a high-load cosmetic industry wastewater (WW) located in Mexico. Discussion is based on a surface response methodology. When using guar, locust bean gum, and Opuntia mucilage, conductivity and turbidity removals as high as 20.1 and 67.8?% were found, respectively. Chemical oxygen demand (COD) removals as high as 38.6?% were observed. The maximum removal efficiency was found for mucilage, with 21.1?mg?COD/mg polymer. At the end of the process, pH was in the range of 5.8?C7.3 for an initial wastewater pH value of 5.6. The production of sludge was very dependent on the WW organic load. An analysis of some metal content in the sludges is presented. From the response surface analysis, it was observed that the parameter which strongly affected the removal of COD, turbidity, oil and greases (O&G), and the amount of sludge including their metal contents was the polymer dose. Only in the case of O&G removal was a combination of dose?Cwastewater organic load responsible for the removals. The values of R ² for the correlation process were between 0.5451 (O&G) and 0.7989 (COD). The p values for the different expressions were between 0.1985 (COD) and 0.7195 (O&G). The values of adequate precisior (AP) indicate how feasible it is to use the surface response analysis (AP?>?4). Most of the analysis indicated that AP?>?4, except in the case of the O&G removal analysis where AP?=?2.9.     

Rice Quality by Spectroscopic Analysis: Precision of Three Spectral Regions

Three types of spectroscopy were used to examine rice quality: near infrared (NIR), Raman, and proton nuclear magnetic resonance (¹H NMR). Samples from 96 rice cultivars were tested. Protein, amylose, transparency, alkali spreading values, whiteness, and degree of milling were measured by standard techniques and the values were regressed against NIR and Raman spectra data. The NMR spectra were used for a qualitative or semiquantitative assessment of the amylose/amylopectin ratio by determining the 1–4 to 1–6 ratio for glucans. Protein can be measured by almost any instrument in any configuration because of the strong relationship between the spectral response and the precision of the reference method. Amylose has an equally strong relationship to the vibrational spectra, but its determination by any reference method is far less precise, resulting in a 10× increase in the standard error of cross‐validation (SECv) or standard error of performance (SEP) with R ² values equal to that of the protein measurement.