WO2021138941A1 - 一种基于拉曼光谱技术的鸡蛋新鲜度无损检测方法 - Google Patents

一种基于拉曼光谱技术的鸡蛋新鲜度无损检测方法 Download PDF

Info

Publication number
WO2021138941A1
WO2021138941A1 PCT/CN2020/072891 CN2020072891W WO2021138941A1 WO 2021138941 A1 WO2021138941 A1 WO 2021138941A1 CN 2020072891 W CN2020072891 W CN 2020072891W WO 2021138941 A1 WO2021138941 A1 WO 2021138941A1
Authority
WO
WIPO (PCT)
Prior art keywords
egg
raman spectrum
partial
raman
freshness
Prior art date
Application number
PCT/CN2020/072891
Other languages
English (en)
French (fr)
Inventor
谢云飞
刘宇良
姚卫蓉
于航
郭亚辉
成玉梁
张颖
张亦弛
Original Assignee
江南大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 江南大学 filed Critical 江南大学
Priority to US17/214,809 priority Critical patent/US20210247318A1/en
Publication of WO2021138941A1 publication Critical patent/WO2021138941A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/65—Raman scattering
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/02—Food
    • G01N33/08—Eggs, e.g. by candling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2201/00—Features of devices classified in G01N21/00
    • G01N2201/02—Mechanical
    • G01N2201/022—Casings
    • G01N2201/0221—Portable; cableless; compact; hand-held
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2201/00—Features of devices classified in G01N21/00
    • G01N2201/12—Circuits of general importance; Signal processing
    • G01N2201/129—Using chemometrical methods

Definitions

  • the invention relates to the field of food detection, in particular to a non-destructive detection method for egg freshness based on Raman spectroscopy technology.
  • eggs can provide the human body with essential nutrients such as protein, fat, minerals and vitamins.
  • essential nutrients such as protein, fat, minerals and vitamins.
  • eggs are particularly prone to corruption in the links of production, processing, sales and circulation, which will not only harm the economic interests of producers, but even endanger the health of consumers, the study of egg freshness has important value and significance. .
  • the detection indicators for egg freshness mainly include Huff units, protein pH value, air chamber diameter, air chamber height, etc., but these indicators need to pass destructive testing to obtain.
  • non-destructive testing has been applied to all aspects.
  • the non-destructive testing techniques applied to egg freshness mainly include infrared spectroscopy, hyperspectral, electronic nose, sonic pulse, machine vision technology, impedance method, etc. There is no non-destructive testing by Raman spectroscopy.
  • the invention establishes a connection between the changed egg surface Raman spectrum and the simultaneously changing physical and chemical indexes of egg freshness, so as to achieve the effect of using the collected Raman spectrum to predict the physical and chemical indexes of freshness.
  • Raman spectroscopy has the advantages of simplicity and speed.
  • Raman spectroscopy can provide information on the vibration frequency of material molecules.
  • the present invention uses Raman spectroscopy as a new detection method for egg freshness. Non-destructive testing.
  • the collected Raman spectrum not only contains the information of the sample itself, but also some other irrelevant information, such as fluorescence background, noise, and stray light. Therefore, it is necessary to preprocess the collected Raman spectra of eggs to filter out the effective spectral information and reduce the interference of irrelevant signals on the results.
  • This patent uses 8 pre-processing methods, namely, curve smoothing (SG), normalization (NL), first-order derivative (1st Der-SG), second-order derivative (2nd Der-SG), and baseline correction (BL). ), standard normal variable transformation (SNV), multivariate scattering correction (MSC), noise reduction (Denoise).
  • SNV standard normal variable transformation
  • MSC multivariate scattering correction
  • Denoise noise reduction
  • the purpose of the present invention is to provide a method for non-destructive detection of egg freshness based on Raman spectroscopy technology, the method is to use Raman spectroscopy information and four physical and chemical indexes of freshness of egg samples to establish partial least squares regression models;
  • the Raman spectrum information is the Raman spectrum intensity of the egg shell surface in the range of 100-3000 cm -1;
  • the physical and chemical indexes of freshness include Huff units, protein pH, air cell diameter and air cell height.
  • the process of establishing the partial least squares regression model includes the following steps:
  • the Raman spectrum information further includes preprocessing, and the preprocessing method includes any one or more of the following combinations: curve smoothing (SG), normalization (NL) , First-order derivative (1st Der), second-order derivative (2nd Der), baseline correction (BL), standard normal variable transformation (SNV), multivariate scattering correction (MSC), noise reduction (Denoise).
  • SG curve smoothing
  • NL normalization
  • NL First-order derivative
  • 2nd Der second-order derivative
  • BL baseline correction
  • SNV standard normal variable transformation
  • MSC multivariate scattering correction
  • Denoise noise reduction
  • the preprocessing method is preferably the second derivative (2nd Der).
  • the preprocessing method is preferably the second derivative (2nd Der).
  • the preprocessing method when constructing the partial least square regression model of Raman spectrum information and the diameter of the gas cell, is preferably the first derivative.
  • the preprocessing method when constructing a partial least squares regression model of Raman spectrum information and the height of the gas cell, is preferably the first derivative (1st Der).
  • the methods for measuring the reference values of the four physical and chemical indicators are as follows:
  • the collection of Raman spectroscopy information includes collecting at any one of three different detection parts, or detecting and averaging at multiple parts; the three different detection parts are respectively For the top of the egg, the bottom of the egg shell and the waist of the egg; take two points for each part and repeat the collection 3 times to calculate the average value.
  • the top of the egg refers to the position of the tip of the egg; the bottom of the egg refers to the position of the head of the egg; and the waist of the egg refers to the middle line of the egg.
  • the Raman spectrum information preferably collects the Raman spectrum intensity of the top eggshell surface of the egg in the range of 100-3000 cm -1.
  • the partial least squares regression model specifically includes the following steps:
  • the egg sample is stored in a constant temperature and humidity incubator, and data collection is performed every few days.
  • the Raman spectrum of the eggshell surface is collected by a portable Raman instrument.
  • the Raman spectrum information acquisition parameters are: excitation wavelength 785nm, acquisition wavelength band 100-3000cm -1 , integration time 5s, scan 3 times, and the distance between the probe and the eggshell surface is 6mm.
  • the collection of the Raman spectrum intensity is to take the Raman spectrum intensities collected by the three detection parts, and take the average value respectively as the representative spectrum intensity of each part; or the three Take the average of all Raman spectra collected by the detection site as the overall average spectrum; respectively establish models between the representative Raman spectra of three different detection sites and the overall average Raman spectrum and four physical and chemical indicators.
  • the partial least squares method is used for modeling, and 80% of the samples are used as the calibration set, and 20% of the samples are used as the prediction set.
  • the invention provides a novel method for non-destructive detection of egg freshness, and provides a new research direction for non-destructive detection of egg freshness.
  • the correlation coefficients of the best prediction models established on the basis of existing data in the present invention are all above 0.8.
  • the correlation coefficients of the model for Huff units, protein pH value and air cell diameter can reach above 0.9, which has good prediction performance and can reach By collecting the eggshell surface spectrum to predict the effect of egg freshness, and expanding the spectrum database through subsequent experiments, the performance of the model can be further improved.
  • the Raman spectroscopy method adopted by the present invention has the advantages of simplicity, rapidity, and on-site in-situ detection.
  • Fig. 1 is a flowchart of a method for non-destructive detection of egg freshness based on Raman spectroscopy technology provided by an embodiment of the present invention.
  • Fig. 2 is a schematic diagram of the overall average Raman spectrum of an egg sample provided by an embodiment of the present invention.
  • HU is the Huff unit
  • H is the protein height, in mm
  • G is the mass of the whole egg, in g.
  • Protein pH Separate the protein into a beaker, stir evenly with a glass rod, and measure the pH value of the obtained protein with a pH meter.
  • Air chamber diameter and air chamber height After breaking the egg, keep the complete air chamber. Use a vernier caliper to measure the air chamber diameter and air chamber height from the inside of the egg. The unit is mm.
  • Egg samples First buy a batch of the freshest eggs.
  • the egg varieties selected in this example are Hailan brown eggs, the chicken age is 200 days, and the eggs are stored in a constant temperature and humidity incubator at 20°C and 40% RH. , Take out a part of the eggs every 3d for data collection.
  • the Raman spectrum prediction model is a partial least squares regression model established by partial least squares regression. 80% of the samples in the test sample are used as the calibration set, and 20% of the samples are used as the prediction set (wherein the calibration set is The sample used to establish the model structure and parameters; the prediction set is the sample used to evaluate the robustness and predictive ability of the model). It is necessary to ensure that the data range of the prediction set is within the data range of the calibration set, and the average value of the calibration set and the prediction set are not different Large, the diversity data of the embodiment is shown in Table 1.
  • the average value of the Raman spectra collected from all parts of the egg is taken as the overall average spectrum, and a full-band PLSR model is established between the four physical and chemical indicators. The results are shown in Table 2.
  • Rc the correlation coefficient of the calibration set
  • RMSEC the root mean square error of the calibration set
  • Rp the correlation coefficient of the prediction set
  • RMSEP the root mean square error of the prediction set.
  • Example 2 The influence of Raman spectroscopy with different processing methods on modeling
  • Example 1 the average value of all Raman spectra collected from the three detection positions was taken as the representative spectrum of the whole, and the overall average Raman spectra were respectively subjected to the following preprocessing: curve smoothing (SG), normalization ( NL), first-order derivative (1st Der-SG), second-order derivative (2nd Der-SG), baseline correction (BL), standard normal variable transformation (SNV), multiple scattering correction (MSC), noise reduction (Denoise) .
  • curve smoothing SG
  • NL normalization
  • NL first-order derivative
  • 2nd Der-SG baseline correction
  • SNV standard normal variable transformation
  • MSC multiple scattering correction
  • Denoise noise reduction
  • the above modeling analysis is performed.
  • the optimal Raman spectrum preprocessing method for each physical and chemical index model is obtained.
  • the best pretreatment method of Raman spectroscopy is the second-order derivation
  • the best pretreatment method of Raman spectroscopy of the air cell diameter and the air cell height is the first-order derivation.
  • the modeling effect may be better.
  • the top of the egg refers to the position of the tip of the egg; the bottom of the egg refers to the position of the head of the egg; and the waist of the egg refers to the middle line of the egg.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Food Science & Technology (AREA)
  • Data Mining & Analysis (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Theoretical Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Medicinal Chemistry (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Operations Research (AREA)
  • Algebra (AREA)
  • Evolutionary Biology (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)

Abstract

一种基于拉曼光谱技术的鸡蛋新鲜度无损检测方法，利用拉曼光谱信息和鸡蛋样品的四项新鲜度理化指标分别建立偏最小二乘回归模型；其中，拉曼光谱信息是鸡蛋蛋壳表面在100-3000cm -1范围内的拉曼光谱强度；新鲜度理化指标包括哈夫单位、蛋白pH、气室直径和气室高度。通过对鸡蛋蛋壳表面拉曼光谱的采集，与对应鸡蛋内部新鲜度的理化指标进行建模分析，模型可达到较好的预测性能，从而达到通过采集拉曼光谱无损预测鸡蛋新鲜度的效果。

Description

一种基于拉曼光谱技术的鸡蛋新鲜度无损检测方法 技术领域
本发明涉及食品检测领域，具体涉及一种基于拉曼光谱技术的鸡蛋新鲜度无损检测方法。
背景技术
鸡蛋作为一种高营养的食物，可以为人体提供必需的蛋白质、脂肪、矿物质和维生素等营养物质。但是由于鸡蛋在生产、加工、销售和流通的环节中，特别容易腐败变质，不仅会损害生产者的经济利益，甚至危害消费者的身体健康，所以对于鸡蛋新鲜度的研究具有重要的价值与意义。
目前对于鸡蛋新鲜度的检测指标主要有哈夫单位、蛋白pH值、气室直径、气室高度等，但这些指标都需要通过有损检测才能获得。随着科技的进步，无损检测被应用到各个方面。目前应用于鸡蛋新鲜度的无损检测技术主要有红外光谱、高光谱、电子鼻、声波脉冲、机器视觉技术、阻抗法等，暂未有通过拉曼光谱进行无损检测。
发明内容
本发明根据变化的鸡蛋表面拉曼光谱与同时变化的鸡蛋新鲜度理化指标之间建立联系，以期达到用采集的拉曼光谱预测新鲜度理化指标的效果。拉曼光谱作为一种适合现场检测的方法，具有简便、快速等优点，同时拉曼光谱可以提供物质分子振动频率的信息，本发明采用拉曼光谱作为一种新型检测方法用于鸡蛋新鲜度的无损检测。
在建模过程中，由于采集的拉曼光谱不仅包含了样本自身的信息，还包含了一些其它的无关信息，如荧光背景、噪音和杂散光等。所以需要对采集到的鸡蛋拉曼光谱图进行预处理，从而筛选出其中有效的光谱信息，降低无关信号对结果的干扰。本专利中共采用了8种前处理方法，分别为曲线平滑(SG)、归一化(NL)、一阶导(1st Der-SG)、二阶导(2nd Der-SG)、基线校正(BL)、标准正态变量变换(SNV)、多元散射校正(MSC)、降噪(Denoise)。另外，本发明经过研究探索，在不同部位均设置检测点采集拉曼光谱较为繁琐，于是本发明对不同采集部位的光谱建模效果进行了比较，采集部位分别为顶部、底部和腰线。
本发明的目的是提供一种基于拉曼光谱技术无损检测鸡蛋新鲜度的方法，所述方法是利用拉曼光谱信息和鸡蛋样品的四项新鲜度理化指标分别建立偏最小二乘回归模型；
所述拉曼光谱信息是鸡蛋蛋壳表面在100-3000cm -1范围内的拉曼光谱强度；
所述新鲜度理化指标包括哈夫单位、蛋白pH、气室直径和气室高度。
在本发明的一种实施方式中，所述偏最小二乘回归模型的建立过程包括如下步骤：
(1)采集鸡蛋样品在100-3000cm -1范围内的拉曼光谱强度；并分别测得相应四项理化指标的参考值；
(2)然后将采集得到的拉曼光谱强度分别与四项理化指标的参考值构建偏最小二乘回归模型。
在本发明的一种实施方式中，所述拉曼光谱信息还包括进行预处理，所述预处理的方式包括如下任意一种或多种组合：曲线平滑(SG)、归一化(NL)、一阶导(1st Der)、二阶导(2nd Der)、基线校正(BL)、标准正态变量变换(SNV)、多元散射校正(MSC)、降噪(Denoise)。
在本发明的一种实施方式中，当构建拉曼光谱信息与哈夫单位的偏最小二乘回归模型时，预处理的方式优选二阶导(2nd Der)。
在本发明的一种实施方式中，当构建拉曼光谱信息与蛋白pH的偏最小二乘回归模型时，预处理的方式优选二阶导(2nd Der)。
在本发明的一种实施方式中，当构建拉曼光谱信息与气室直径的偏最小二乘回归模型时，预处理的方式优选一阶导(1st Der)。
在本发明的一种实施方式中，当构建拉曼光谱信息与气室高度的偏最小二乘回归模型时，预处理的方式优选一阶导(1st Der)。
在本发明的一种实施方式中，四项理化指标的参考值的测定方法分别如下：
利用蛋品分析仪测定鸡蛋的哈夫单位；利用pH计测定蛋白pH；利用游标卡尺测量气室直径与气室高度。
在本发明的一种实施方式中，拉曼光谱信息的采集包括在三个不同的检测部位中任意一种进行采集，或者在多种部位检测取平均值；所述三个不同的检测部位分别为鸡蛋顶部、鸡蛋壳底部和鸡蛋腰部；每个部位取两点重复采集3次计平均值。
在本发明的一种实施方式中，鸡蛋顶部是指鸡蛋尖头位置；鸡蛋底部是指鸡蛋顿头位置；鸡蛋腰部是指鸡蛋中间线部位。
在本发明的一种实施方式中，所述拉曼光谱信息优选采集鸡蛋的顶部蛋壳表面在100-3000cm -1范围内的拉曼光谱强度。
在本发明的一种实施方式中，所述偏最小二乘回归模型具体包括如下步骤：
(1)数据采集：检测每个鸡蛋的拉曼光谱以及对应新鲜度理化指标；
(2)数据建模：将采集的拉曼光谱与哈夫单位、蛋白pH、气室直径和气室高度这四项新鲜度理化指标分别建模，根据所建的模型，可以通过采集的光谱直接获得鸡蛋内部新鲜度 理化指标的数值，达到无损、快速、准确检测鸡蛋新鲜度的目的。
在本发明的一种实施方式中，所述鸡蛋样品是保存恒温恒湿培养箱中，每隔几天进行数据的采集。
在本发明的一种实施方式中，通过便携式拉曼仪采集蛋壳表面的拉曼光谱。
在本发明的一种实施方式中，所述拉曼光谱信息的采集参数为：激发波长785nm，采集波段为100-3000cm -1，积分时间5s，扫描3次，探头与蛋壳表面距离6mm。
在本发明的一种实施方式中，所述拉曼光谱强度的采集是将三个检测部位采集到的拉曼光谱强度，分别各自取平均值作为每个部位的代表光谱强度；或者将三个检测部位采集的所有拉曼光谱取平均值作为整体的平均光谱；分别将三个不同检测部位的代表拉曼光谱以及整体的平均拉曼光谱与四种理化指标之间建立模型。
在本发明的一种实施方式中，采用偏最小二乘法建模，将80％的样品作为校正集，20％的样品作为预测集。
有益效果：
本发明提供了一种新型的无损检测鸡蛋新鲜度的方法，为鸡蛋新鲜度的无损检测提供了新的研究方向。本发明中基于已有数据建立的最佳预测模型相关系数均在0.8以上，其中对于哈夫单位、蛋白pH值和气室直径的模型相关系数可达0.9以上，具有较好的预测性能，可达到通过采集蛋壳表面光谱预测鸡蛋新鲜度的效果，通过后续实验对光谱数据库进行扩大，模型的性能可进一步提升。同时，本发明采用的拉曼光谱方法具有简便、快速、可现场原位检测等优点，相比于红外光谱还可避免水分子对光谱结果的干扰，通过数据库的扩大，更容易走向实际应用。另外，此方法在其他食品的新鲜度检测领域也可能适用，可进行相关的研究。
附图说明
图1本发明实施例提供的基于拉曼光谱技术的鸡蛋新鲜度无损检测方法流程图。
图2本发明实施例提供的鸡蛋样本的整体平均拉曼光谱示意图。
具体实施方式
为了更好理解本发明，下面结合实施例对本发明做进一步说明，但本发明要求保护的范围并不仅仅局限于实施例表述的范围。
下面对本发明的操作流程作详细的描述。
哈夫单位：根据蛋重和浓厚蛋白高度，按公式计算出来的值，用来衡量蛋的新鲜程度，公式为：HU＝100×lg(H+7.57-1.7×G 0.37)。式中：HU为哈夫单位；H为蛋白高度，单位为mm；G为整颗鸡蛋质量，单位为g。
蛋白pH：将蛋白分离至烧杯中，玻璃棒搅匀，利用pH计测定得到的蛋白pH值。
气室直径、气室高度：将鸡蛋打破后，保留完整的气室，利用游标卡尺从蛋内部分别测量得到气室直径和气室高度，单位均为mm。
实施例1构建拉曼光谱预测模型
(1)鸡蛋样品：首先购置一批最新鲜的鸡蛋，本实施例中选用的鸡蛋品种为海兰褐鸡蛋，鸡龄为200d，鸡蛋保存在20℃、40％RH的恒温恒湿培养箱中，每3d取出一部分鸡蛋进行数据的采集。
(2)采集每个样品的拉曼光谱：采用便携拉曼仪采集样本的顶部、底部、腰线拉曼光谱数据，每个部位选择两点，每点重复测试3次。拉曼检测参数为：积分时间5s，扫描次数3次，采集波段为全波段100-3000cm -1，探头与蛋壳表面的距离6mm；测定相应的拉曼光谱信息。
(3)检测每个样品的各理化指标参考值：通过蛋品分析仪检测鸡蛋的哈夫单位；利用pH计测定蛋白pH；利用游标卡尺测量气室直径与气室高度；分别得到哈夫单位、蛋白pH、气室直径和气室高度的参考值；
(4)数据建模分析：结合化学计量学，利用测得的鸡蛋在全波段100-3000cm -1的拉曼光谱强度信息分别与各理化指标参考值建立偏最小二乘回归(PLSR)模型；即为鸡蛋新鲜度理化指标的拉曼光谱预测模型。
所述的拉曼光谱预测模型为采用偏最小二乘回归法建立的偏最小二乘回归模型，将测试样品中80％的样品作为校正集，20％的样品作为预测集(其中，校正集为用来建立模型结构和参数的样品；预测集为用来评估模型稳健性和预测能力的样品)，需保证预测集数据范围在校正集数据范围内，且校正集和预测集的平均值相差不大，实施例的分集数据见表1。
表1各理化指标的分集统计数据
Figure PCTCN2020072891-appb-000001
Figure PCTCN2020072891-appb-000002
本实施例中，将鸡蛋所有部位采集到的拉曼光谱取平均值作为整体平均光谱，与四种理化指标之间建立全波段PLSR模型，结果见表2。
表2原始拉曼光谱与各理化指标建模结果
Figure PCTCN2020072891-appb-000003
其中，Rc：校正集的相关系数；RMSEC：校正集均方根误差；Rp：预测集的相关系数；RMSEP：预测集均方根误差。
由表2可知，未经预处理的平均拉曼光谱，与蛋白pH值建模后模型的相关系数均可达0.6以上，较好可达0.8以上，说明该模型具有较好的预测性能，另外，与其他三个理化指标建模后模型相关系数也均在0.6以上，若光谱经过预处理后很有可能得到具有更稳定、更佳的预测性能的模型。
实施例2不同处理方式处理拉曼光谱对建模的影响
参照实施例1，将三个检测部位采集的所有拉曼光谱取平均值作为整体的代表光谱，并将整体的平均拉曼光谱分别经过了以下预处理：曲线平滑(SG)、归一化(NL)、一阶导(1st Der-SG)、二阶导(2nd Der-SG)、基线校正(BL)、标准正态变量变换(SNV)、多元散射校正(MSC)、降噪(Denoise)。以哈夫单位值为例，全波段PLSR建模结果见表3。
表3不同处理方式处理拉曼光谱对建模的影响
Figure PCTCN2020072891-appb-000004
进一步的，根据整体平均拉曼光谱与每个理化指标进行如上建模分析，对于仅采用一种前处理方法的模型，得到了每个理化指标模型的最佳拉曼光谱预处理方法。其中哈夫单位和蛋白pH值建模过程中，拉曼光谱最佳预处理方法为二阶导，气室直径和气室高度的拉曼光谱最佳预处理方法为一阶导。当然，如果将几种前处理方式组合起来，建模效果可能会更佳。
实施例3不同拉曼测定部位对建模的影响
将三个检测部位采集到的拉曼光谱分别各自取平均值作为每个部位的代表光谱，再分别将三个不同检测部位的代表拉曼光谱采用前面得到的最佳预处理方法，其他参照实施例1，与四种理化指标之间建立全波段PLSR模型，结果见表4。
表4不同部位代表拉曼光谱与各理化指标建模结果
Figure PCTCN2020072891-appb-000005
其中，鸡蛋顶部是指鸡蛋尖头位置；鸡蛋底部是指鸡蛋顿头位置；鸡蛋腰部是指鸡蛋中间线部位。
由以上结果可知，对于鸡龄为200d的海兰褐鸡蛋，采集顶部拉曼光谱，并采用二阶导或者一阶导对光谱进行预处理，与鸡蛋新鲜度的理化指标建立100-3000cm -1的全波段偏最小二乘回归模型，模型可达到较好的预测性能。其中对于哈夫单位值，Rc为0.944，Rp为0.925；对于蛋白pH值，Rc为0.945，Rp为0.935；对于气室直径，Rc为0.903，Rp为0.915；对于气室高度，Rc为0.828，Rp为0.830。
基于拉曼光谱技术对海兰褐鸡蛋以及其它品种鸡蛋的其它指标(如蛋黄指数等)进行无损检测时，可参照该文所提出的检测方法和检测流程进行操作。
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (11)

  1. 一种基于拉曼光谱技术无损检测鸡蛋新鲜度的方法，其特征在于，所述方法是利用拉曼光谱信息和鸡蛋样品的四项新鲜度理化指标分别建立偏最小二乘回归模型；
    所述拉曼光谱信息是鸡蛋顶部蛋壳表面在100-3000cm -1范围内的拉曼光谱强度；
    所述新鲜度理化指标包括哈夫单位、蛋白pH、气室直径和气室高度；
    其中，当构建拉曼光谱信息与哈夫单位的偏最小二乘回归模型时，预处理的方式为二阶导；当构建拉曼光谱信息与蛋白pH的偏最小二乘回归模型时，预处理的方式为二阶导；当构建拉曼光谱信息与气室直径的偏最小二乘回归模型时，预处理的方式为一阶导；当构建拉曼光谱信息与气室高度的偏最小二乘回归模型时，预处理的方式为一阶导。
  2. 一种基于拉曼光谱技术无损检测鸡蛋新鲜度的方法，其特征在于，所述方法是利用拉曼光谱信息和鸡蛋样品的四项新鲜度理化指标分别建立偏最小二乘回归模型；
    所述拉曼光谱信息是鸡蛋蛋壳表面在100-3000cm -1范围内的拉曼光谱强度；
    所述新鲜度理化指标包括哈夫单位、蛋白pH、气室直径和气室高度。
  3. 根据权利要求1所述的方法，其特征在于，所述偏最小二乘回归模型的建立过程包括如下步骤：
    (1)采集鸡蛋样品在100-3000cm -1范围内的拉曼光谱强度；并分别测定鸡蛋样品的四项理化指标作为参考值；
    (2)然后将采集得到的拉曼光谱强度分别与四项理化指标的参考值构建偏最小二乘回归模型。
  4. 根据权利要求2所述的方法，其特征在于，所述拉曼光谱信息还包括进行预处理，所述预处理的方式包括如下任意一种或多种组合：曲线平滑、归一化、一阶导、二阶导、基线校正、标准正态变量变换、多元散射校正、降噪。
  5. 根据权利要求4所述的方法，其特征在于，当构建拉曼光谱信息与哈夫单位的偏最小二乘回归模型时，预处理的方式为二阶导。
  6. 根据权利要求4所述的方法，其特征在于，当构建拉曼光谱信息与蛋白pH的偏最小二乘回归模型时，预处理的方式为二阶导。
  7. 根据权利要求4所述的方法，其特征在于，当构建拉曼光谱信息与气室直径的偏最小二乘回归模型时，预处理的方式为一阶导。
  8. 根据权利要求4所述的方法，其特征在于，当构建拉曼光谱信息与气室高度的偏最小二乘回归模型时，预处理的方式为一阶导。
  9. 根据权利要求2所述的方法，其特征在于，拉曼光谱信息的采集包括在三个不同的检测部位中任意一种进行采集，或者在多种部位检测取平均值；所述三个不同的检测部位分别 为鸡蛋顶部、鸡蛋底部和鸡蛋腰部。
  10. 根据权利要求2所述的方法，其特征在于，所述拉曼光谱信息为采集鸡蛋的顶部蛋壳表面在100-3000cm -1范围内的拉曼光谱强度。
  11. 根据权利要求2所述的方法，其特征在于，所述拉曼光谱信息的采集参数为：激发波长785nm，采集波段为100-3000cm -1，积分时间5s，扫描3次，探头与鸡蛋的蛋壳表面距离为6mm。
PCT/CN2020/072891 2020-01-09 2020-01-18 一种基于拉曼光谱技术的鸡蛋新鲜度无损检测方法 WO2021138941A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/214,809 US20210247318A1 (en) 2020-01-09 2021-03-27 Non-Destructive Detection of Egg Freshness Based on Raman Spectroscopy

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010023268.2A CN111157511B (zh) 2020-01-09 2020-01-09 一种基于拉曼光谱技术的鸡蛋新鲜度无损检测方法
CN202010023268.2 2020-01-09

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/214,809 Continuation US20210247318A1 (en) 2020-01-09 2021-03-27 Non-Destructive Detection of Egg Freshness Based on Raman Spectroscopy

Publications (1)

Publication Number Publication Date
WO2021138941A1 true WO2021138941A1 (zh) 2021-07-15

Family

ID=70562129

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/072891 WO2021138941A1 (zh) 2020-01-09 2020-01-18 一种基于拉曼光谱技术的鸡蛋新鲜度无损检测方法

Country Status (3)

Country Link
US (1) US20210247318A1 (zh)
CN (1) CN111157511B (zh)
WO (1) WO2021138941A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114578002A (zh) * 2022-03-09 2022-06-03 武汉轻工大学 一种基于鸡蛋容重的鸡蛋品质无损检测方法

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109813494A (zh) * 2019-02-27 2019-05-28 西北农林科技大学 一种基于测量质心无损检测鸡蛋新鲜度的方法
CN112179746B (zh) * 2020-09-24 2021-11-23 江南大学 一种定量评价鸡蛋胶护膜品质的方法
IL280284A (en) * 2021-01-19 2022-08-01 Zvi Nezer A method for optically testing the liquid content of an egg
CN113655050B (zh) * 2021-08-17 2024-04-26 南京富岛信息工程有限公司 一种改进轻质油中痕量原油拉曼光谱检测限的方法
CN113866153A (zh) * 2021-09-27 2021-12-31 江南大学 一种基于拉曼光谱技术的猪肉新鲜度快速检测方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050221506A1 (en) * 2004-03-30 2005-10-06 Intel Corporation Surface modification of metals for biomolecule detection using surface enhanced Raman scattering (SERS)
CN103424394A (zh) * 2013-08-23 2013-12-04 江苏大学 一种冷冻南美白对虾品质的拉曼无损检测方法
CN104597028A (zh) * 2015-01-09 2015-05-06 河南工业大学 一种小麦新鲜度指标的快速检测方法
CN106706596A (zh) * 2017-01-09 2017-05-24 河南工业大学 一种大米新鲜度的拉曼光谱检测方法
CN109001152A (zh) * 2018-10-08 2018-12-14 中国农业大学 基于可见/近红外光谱的禽蛋新鲜度综合指标检测方法
TW201913072A (zh) * 2017-09-01 2019-04-01 鄭經偉 非破壞性光譜檢測蛋的新鮮度之方法
CN208752006U (zh) * 2018-07-23 2019-04-16 南京农业大学 一种基于近红外光谱的鸡蛋新鲜度检测设备

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105548028A (zh) * 2015-12-11 2016-05-04 华中农业大学 禽蛋新鲜度的光纤光谱分级检测装置及其方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050221506A1 (en) * 2004-03-30 2005-10-06 Intel Corporation Surface modification of metals for biomolecule detection using surface enhanced Raman scattering (SERS)
CN103424394A (zh) * 2013-08-23 2013-12-04 江苏大学 一种冷冻南美白对虾品质的拉曼无损检测方法
CN104597028A (zh) * 2015-01-09 2015-05-06 河南工业大学 一种小麦新鲜度指标的快速检测方法
CN106706596A (zh) * 2017-01-09 2017-05-24 河南工业大学 一种大米新鲜度的拉曼光谱检测方法
TW201913072A (zh) * 2017-09-01 2019-04-01 鄭經偉 非破壞性光譜檢測蛋的新鮮度之方法
CN208752006U (zh) * 2018-07-23 2019-04-16 南京农业大学 一种基于近红外光谱的鸡蛋新鲜度检测设备
CN109001152A (zh) * 2018-10-08 2018-12-14 中国农业大学 基于可见/近红外光谱的禽蛋新鲜度综合指标检测方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
ZHAO Q., BAN L., ZHENG J., XU G., NING Z., QU L.: "Potential use of spectroscopic techniques for assessing table eggs and hatching eggs", WORLDS POULTRY SCIENCE JOURNAL, vol. 75, no. 3, 30 September 2019 (2019-09-30), pages 445 - 456, XP009529212, ISSN: 0043-9339, DOI: 10.1017/S0043933919000424 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114578002A (zh) * 2022-03-09 2022-06-03 武汉轻工大学 一种基于鸡蛋容重的鸡蛋品质无损检测方法

Also Published As

Publication number Publication date
US20210247318A1 (en) 2021-08-12
CN111157511B (zh) 2021-07-27
CN111157511A (zh) 2020-05-15

Similar Documents

Publication Publication Date Title
WO2021138941A1 (zh) 一种基于拉曼光谱技术的鸡蛋新鲜度无损检测方法
Su et al. Evaluation of spectral imaging for inspection of adulterants in terms of common wheat flour, cassava flour and corn flour in organic Avatar wheat (Triticum spp.) flour
Ali et al. Rapid and nondestructive techniques for internal and external quality evaluation of watermelons: A review
WO2005111583A1 (ja) 近赤外線分光法による野菜等の成分の非破壊検査法、及び同装置
CN102967578A (zh) 在线获取牛肉样本近红外光谱的方法及其在评价牛肉质量中的应用
CN104965973B (zh) 一种苹果霉心病多因子无损检测判别模型及其建立方法
CN106525759A (zh) 一种基于衰减全反射太赫兹介电谱鉴别蜂蜜品种的方法
Xu et al. Factors influencing near infrared spectroscopy analysis of agro-products: a review
Chang et al. Non-destructive identification of internal watercore in apples based on online Vis/NIR spectroscopy
Nturambirwe et al. Detecting bruise damage and level of severity in apples using a contactless nir spectrometer
CN108169168A (zh) 检测分析水稻籽粒蛋白质含量数学模型及构建方法和应用
CN104255118A (zh) 基于近红外光谱技术的水稻种子发芽率快速无损测试方法
Ferrara et al. The prediction of ripening parameters in Primitivo wine grape cultivar using a portable NIR device
CHEN et al. NIR measurement of specific gravity of potato
Janni et al. Novel near-infrared sampling apparatus for single kernel analysis of oil content in maize
Wang et al. Monitoring model for predicting maize grain moisture at the filling stage using NIRS and a small sample size
CN106772427B (zh) 一种基于连续小波分析建立小麦叶干重定量模型的方法
Liu et al. Prediction of valid acidity in intact apples with Fourier transform near infrared spectroscopy
CN104089883B (zh) 基于激光器阵列的检测装置及小黄鱼存储时间检测方法
CN102636455A (zh) 一种近红外光谱测定双孢蘑菇硬度的方法
Liu et al. Non-destructive detection of single corn seed vigor based on visible/near-infrared spatially resolved spectroscopy combined with chemometrics
Purwanto et al. Measurement of starch and soluble solid content in papaya using near infrared spectroscopy
Li Nondestructive measurement model of apple internal browning based on FT-NIR spectroscopy
Li et al. Non-destructive Detection the Content of Acid Detergent Fiber in Corn Stalk Using NIRS
Zhang Shatang mandarin sugar degree detection based on near infrared spectrum

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20911511

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20911511

Country of ref document: EP

Kind code of ref document: A1