Near-infrared system for pea component content detection and quality grading

Abstract

A near-infrared system for pea component content detection and quality grading comprises a detection module, an analysis and judgment module and a storage module, and the detection module, the analysis and judgment module and the storage module can be arranged on a portable near-infrared spectrometer. According to the invention, the pea near-infrared component detection model in the detection module only needs to carry out near-infrared detection on the pea sample, the basic component content of the pea sample can be obtained through the analysis and judgment module based on the obtained near-infrared reflection spectrum, and the component content is graded according to the pea quality grade standard; whether pea varieties are suitable for being used as starch and protein processing raw materials can be judged. According to the system and the technology, the detection efficiency of the basic components of the pea samples is effectively improved, and the pea samples cannot be damaged, so that the pea samples can be reutilized and can be continuously stored to be detected again when problems occur in subsequent production, and multiple batches of pea samples can be rapidly detected.

Description

Technical Field

[0001] The invention belongs to the field of pea component detection, in particular to a near-infrared system for pea component content detection and quality grading. Background Art

[0002] Peas are one of the most common crops at present. Peas have a high protein content. Pea protein is also one of the most common protein powders on the market. Since the ingredients of peas of different varieties from different origins vary greatly, when deep processing peas (such as extracting pea protein) is carried out, it is necessary to test the ingredients of peas of different varieties harvested from different origins to determine the content of their basic ingredients, so as to determine whether they meet the standards for deep processing and what kind of deep processing technology they are suitable for. However, the current chemical method detection requires separate detection operations, which is relatively complicated, and the pea samples need to be destroyed for detection. At the same time, batch detection leads to long detection time, low detection efficiency and the inability to preserve pea samples. Summary of the invention

[0003] The invention provides a near infrared system for detecting the content of pea components and grading the quality, so as to solve the defects in the prior art.

[0004] The present invention is achieved through the following technical solutions:

[0005] A near-infrared system for detecting the content of pea ingredients and grading their quality, comprising a detection module, an analysis and judgment module and a storage module, wherein the detection module, the analysis and judgment module and the storage module can all be arranged in a portable near-infrared spectrometer;

[0006] A near-infrared system for detecting the content of pea ingredients and grading the quality as described above, wherein the detection module is used to perform near-infrared detection on peas and obtain the near-infrared reflectance spectrum of peas;

[0007] In the above-mentioned near-infrared system for detecting the content of pea components and grading the quality, the analysis and judgment module can analyze and judge the near-infrared reflectance spectrum of peas obtained by the detection module through the near-infrared component detection model of peas to obtain the specific content of pea components, grade the content of the components, and finally output the content of various components of peas and the corresponding grades;

[0008] In the above-mentioned near-infrared system for detecting the content of pea components and grading their quality, the storage module is used to store the results obtained by the analysis and judgment module and output and display them through the display screen of the portable near-infrared spectrometer.

[0009] A near infrared system for detecting the content of pea ingredients and grading their quality as described above, wherein the ingredients in step 2 include protein, starch, water, crude fiber and crude fat;

[0010] A near infrared system for detecting the content and grading the quality of pea components as described above, wherein the first-grade protein content is greater than or equal to 25%, the second-grade content is 20-25%, and the third-grade content is less than or equal to 20%;

[0011] A near infrared system for detecting the content of pea components and grading their quality as described above, wherein the first-grade starch content is greater than or equal to 55, the second-grade content is between 45-55%, and the third-grade content is less than or equal to 45%;

[0012] A near-infrared system for detecting the content of pea components and grading their quality as described above, wherein the first-level moisture content is greater than or equal to 13%, the second-level content is between 10-13%, and the third-level content is less than or equal to 10%;

[0013] A near-infrared system for detecting the content and grading the quality of pea components as described above, wherein the crude fiber content of the first grade is greater than or equal to 7%, the content of the second grade is 5-7%, and the content of the third grade is less than or equal to 5%;

[0014] In the above-mentioned near-infrared system for detecting the content of pea components and grading their quality, the first-grade crude fat content is less than or equal to 5%, the second-grade content is between 5-9%, and the third-grade content is greater than or equal to 9%.

[0015] The near infrared system for detecting the content and grading the quality of pea components as described above, and the method for constructing the near infrared component detection model of peas include the following steps:

[0016] Step 1: Select peas of different varieties from different origins as samples and screen out impurities and damaged peas before labeling the samples;

[0017] Step 2: All pea samples are tested using a portable near-infrared spectrometer to obtain near-infrared reflectance spectra of the corresponding pea samples;

[0018] Step 3: using chemical methods to determine the basic component contents of all pea samples obtained from step 2 with near infrared reflectance spectra;

[0019] Step 4: Use the partial least squares regression correction method to establish a pea near-infrared component detection model based on the near-infrared reflectance spectrum of the pea sample and the basic component content of the pea sample.

[0020] In the above-mentioned near-infrared system for detecting the content of pea components and grading the quality, the screened pea samples in step 1 are placed in a refrigerator and stored at a temperature of 4° C.

[0021] As described above, in a near-infrared system for detecting component content and grading quality of peas, before the portable near-infrared spectrometer obtains the near-infrared reflectance spectrum, the portable near-infrared spectrometer and the pea sample are placed in the same environment for 24 hours to keep the pea sample environment consistent with the operating environment of the portable near-infrared spectrometer.

[0022] In the above-mentioned near-infrared system for detecting the content of pea components and grading the quality, the width, length and thickness of the pea samples after screening in the step 1 are repeatedly analyzed five times by (Shanghai Tool Factory Co., Ltd., Shanghai, China), and the average value is taken (refer to GB 10385-89); the weight of 100 grains of the screened pea samples is weighed by an analytical balance, and the weighing is repeated three times to take the average value (refer to SN 0798-1999).

[0023] The near-infrared system for detecting the content of pea components and grading the quality as described above, wherein the basic components of the pea samples include starch, crude protein, crude fiber, water and crude fat, and the chemical analysis of the basic components of each pea sample is repeated three times, and the results are averaged for data analysis; the starch content of the pea samples is measured by a fully automatic starch detector (FOSS, Denmark) (reference GB 5009.9-2016 second method); the crude protein content of the pea sample was determined by the Kjeldahl method (reference GB 5009.5-2016 first method), and the conversion factor was 5.46 using a 2300 nitrogen analyzer (FOSS, Denmark); the moisture content of the pea samples was detected by drying method (reference GB / T21305-2007); the crude fat content of the pea samples was detected by automatic Soxhlet total fat analyzer (FOSS, Denmark) (reference GB 5009.6-2016 second method); the crude fiber of the pea sample was determined by a semi-automatic fiber measuring instrument (FOSS, Denmark) determination (reference GB / T5009.10-2003).

[0024] As described above, a near-infrared system for detecting the content of pea components and grading the quality, the spectral range of the portable near-infrared spectrometer is 908-1676nm, the variable is 90, the sampling interval is 6nm, the light source is a 10W halogen lamp, and the spectral integration time is 12.7ms. In the process of spectrum acquisition using the portable near-infrared spectrometer, the room temperature must be maintained at 25°C. When collecting the spectrum for each pea sample, the temperature of the portable near-infrared spectrometer must be kept consistent, otherwise the information carried by the spectrum will change, causing baseline drift. Therefore, when using the portable near-infrared spectrometer, it must first be preheated at room temperature for 1h after turning on the machine, and the portable near-infrared spectrometer can be measured after the temperature reaches a constant level. The pea sample is placed in a sample cup (50mm high; 51mm in diameter), and gently shaken to ensure that the pea seeds are evenly distributed. Each pea sample is scanned 5 times, and the sample cup is rotated a certain angle each time it is scanned. Repeat the sample loading and scanning three times to obtain the average value. Due to the influence of sample heterogeneity and distribution heterogeneity, the spectra of different parts of a sample are different. Therefore, multiple mixing and multiple scanning are required to obtain the average spectrum as the original spectrum to reduce the error.

[0025] As described above, a near-infrared system for detecting the content and quality grading of pea components, the near-infrared component detection model of peas is established using The Unscrambler X 10.3 software (CAMO, Norway) and MATLAB R2021a software (MathWorks, USA).

[0026] In the above-mentioned near-infrared system for detecting the component content and grading the quality of peas, in the step 4, the principal component analysis program (PCA) is written by MATLAB R2021a software in combination with the Mahalanobis distance program to screen out abnormal samples from the near-infrared reflectance spectra and basic component content values ​​of all pea samples, and the near-infrared spectral data of the remaining qualified sample sets are preprocessed by the near-infrared spectroscopy using The Unscrambler X 10.3 software, so as to establish an accurate and robust pea near-infrared component detection model for the basic component content of peas.

[0027] In the above-mentioned near-infrared system for detecting the content and grading the quality of pea components, the fourth step uses the Kennard-stone algorithm (3:1) to divide the calibration set and the validation set samples during the process of establishing the pea near-infrared component detection model.

[0028] In the above-mentioned near-infrared system for detecting component content and grading quality of peas, the number of pea samples selected is greater than or equal to 90.

[0029] The advantages of the present invention are: the present invention only needs to perform near-infrared detection on the pea sample through the pea near-infrared component detection model in the analysis and judgment module, and the basic component content of the pea sample can be obtained through the obtained near-infrared reflectance spectrum and the component content can be graded, thereby effectively improving the detection efficiency of the basic component of the pea sample, and the detection process will not destroy the pea sample, so the pea sample can be reused as a raw material, and can also be continuously stored and re-tested when problems occur in subsequent production, and multiple batches of pea samples can be quickly tested, effectively shortening the time for detecting the basic components of peas, greatly improving the detection efficiency of the basic components of peas, and being able to screen out raw materials suitable for starch and protein processing, providing support for the selection of raw materials for enterprise production, and being able to detect without destroying the pea sample, ensuring the normal progress of subsequent processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0031] Figure 1 is a schematic diagram of a portable near-infrared spectrometer including the system of the present invention;

[0032] Figure 2 is a schematic diagram of a pea sample of the present invention;

[0033] Figure 3 It is a schematic diagram of the original spectrum of the pea sample of the present invention. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] like Figure 1 As shown, a near-infrared system for detecting the content of pea ingredients and grading the quality includes a detection module, an analysis and judgment module, and a storage module. The detection module, the analysis and judgment module, and the storage module can all be arranged in a portable near-infrared spectrometer;

[0036] Preferably, the detection module is used to perform near-infrared detection on peas and obtain the near-infrared reflectance spectrum of peas;

[0037] Preferably, the analysis and judgment module can analyze and judge the pea near-infrared reflectance spectrum obtained by the detection module through the pea near-infrared component detection model to obtain the specific pea component content and grade the component content, and finally output the content of various pea components and the corresponding grades;

[0038] Preferably, the storage module is used to store the results obtained by the analysis and judgment module and output and display them through the display screen of the portable near-infrared spectrometer.

[0039] Preferably, the ingredients in step 2 include protein, starch, water, crude fiber and crude fat;

[0040] Preferably, the first-grade protein content is greater than or equal to 25%, the second-grade content is between 20-25%, and the third-grade content is less than or equal to 20%;

[0041] Preferably, the first grade starch content is greater than or equal to 55%, the second grade content is 45-55%, and the third grade content is less than or equal to 45%;

[0042] Preferably, the first level of moisture content is greater than or equal to 13%, the second level of moisture content is 10-13%, and the third level of moisture content is less than or equal to 10%;

[0043] Preferably, the crude fiber first grade content is greater than or equal to 7%, the second grade content is 5-7%, and the third grade content is less than or equal to 5%;

[0044] Preferably, the first grade crude fat content is less than or equal to 5%, the second grade content is 5-9%, and the third grade content is greater than or equal to 9%.

[0045] Preferably, the near-infrared characteristic peak range of pea starch is 800 to 1100 nm.

[0046] 800-1100nm: The characteristic peaks in this range usually correspond to the vibration of COC bonds and CH bonds in starch;

[0047] Preferably, the near-infrared characteristic peak range of pea protein is from 1100 to 1600 nm.

[0048] 1100-1300nm: In this range, CH vibration, NH vibration and CN vibration in proteins can be observed; 1300-1500nm: The characteristic peaks in this range usually correspond to the vibrations of carboxyl (COOH) and hydroxyl (OH) groups in proteins;

[0049] 1500-1600nm: This range usually includes the C=O vibrations in proteins.

[0050] Preferably, the near-infrared characteristic peak range of pea crude fiber is from 1600 to 2500 nm.

[0051] 1600-1800nm: This range usually includes the C=O vibration in cellulose;

[0052] 1800-2500nm: In this range, CO vibrations in cellulose can be observed.

[0053] Preferably, the near-infrared characteristic peak range of pea crude fat is: 4000 to 5200 cm^-1.

[0054] 4000-4200cm^-1: In this range, CH stretching vibrations in fat molecules can be observed;

[0055] 4200-4400 cm^-1: The characteristic peaks in this range usually correspond to the C=O vibrations in fats;

[0056] 4400-4600cm^-1: This range usually includes the OH vibration in fat;

[0057] 4600-4800cm^-1: In this range, CO vibrations in fat can be observed;

[0058] 4800-5200cm^-1: In this range, there are also some characteristic peaks corresponding to the vibration of fatty acid groups.

[0059] Preferably, the near-infrared characteristic peak range of pea moisture is 5200 to 5800 cm^-1.

[0060] 5200-5400cm^-1: In this range, the vibration of the OH group in the water molecule can be observed;

[0061] 5600-5800 cm^-1: Characteristic peaks in this range usually correspond to HOH bending vibrations in water molecules.

[0062] The near-infrared reflectance spectrum of peas can be accurately analyzed through the characteristic peak areas of protein, starch, moisture, crude fiber and crude fat to ensure the correctness of the test results.

[0063] Preferably, the method for constructing the pea near-infrared component detection model comprises the following steps:

[0064] Step 1: Select peas of different varieties from different origins as samples and screen out impurities and damaged seeds and then label the samples (e.g. Figure 2 shown);

[0065] Step 2: All pea samples are tested using a portable near-infrared spectrometer to obtain near-infrared reflectance spectra of the corresponding pea samples;

[0066] Step 3: using chemical methods to determine the basic component contents of all pea samples obtained from step 2 with near infrared reflectance spectra;

[0067] Step 4: Use the partial least squares regression correction method to establish a pea near-infrared component detection model based on the near-infrared reflectance spectrum of the pea sample and the basic component content of the pea sample.

[0068] Preferably, partial least squares regression is a regression modeling method for multiple dependent variables on multiple independent variables. This method is used to solve the interdependence between two sets of multiple related variables, and study the use of one set of variables (independent variables or predictor variables) to predict another set of variables (dependent variables or response variables) (Hao Huizhen, 2021). When the number of two sets of variables is large and linearly correlated, and the number of observed data is small, it is suitable to use partial least squares regression to establish a model. Compared with traditional multiple linear regression, it effectively solves the problem of multicollinearity between variables, has the advantages of simple calculation, high prediction accuracy, and easy qualitative interpretation. The model constructed by it can more accurately identify effective information.

[0069] Specifically, the screened pea samples in step 1 of this embodiment were placed in a refrigerator and stored at 4°C.

[0070] Specifically, before the portable near-infrared spectrometer described in this embodiment obtains the near-infrared reflectance spectrum, the portable near-infrared spectrometer and the pea sample are placed in the same environment for 24 hours to keep the pea sample environment consistent with the operating environment of the portable near-infrared spectrometer.

[0071] More specifically, the width, length and thickness of the pea samples screened in step 1 of this embodiment were analyzed five times by (Shanghai Tool Factory Co., Ltd., Shanghai, China) and the average value was taken (refer to GB 10385-89); the 100-grain weight of the screened pea samples was weighed by an analytical balance, and the weighing was repeated three times to take the average value (refer to SN0798-1999).

[0072] More specifically, the basic components of the pea samples described in this embodiment include starch, crude protein, crude fiber, water and crude fat. The chemical analysis of the basic components of each pea sample was repeated three times, and the results were averaged for data analysis. The starch content of the pea samples was measured using a fully automatic starch detector (FOSS, Denmark) (reference GB 5009.9-2016 second method); the crude protein content of the pea sample was determined by the Kjeldahl method (reference GB5009.5-2016 first method), and the conversion factor was 5.46 using a 2300 nitrogen analyzer (FOSS, Denmark); the moisture content of the pea samples was detected by drying method (reference GB / T 21305-2007); the crude fat content of the pea samples was determined by automatic Soxhlet total fat analyzer (FOSS, Denmark) (reference GB 5009.6-2016 second method); the crude fiber of the pea sample was determined by a semi-automatic fiber measuring instrument (FOSS, Denmark) determination (reference GB / T 5009.10-2003).

[0073] More specifically, the spectral range of the portable near-infrared spectrometer described in this embodiment is 908-1676nm, the variable is 90, the sampling interval is 6nm, the light source is a 10W halogen lamp, and the spectral integration time is 12.7ms. In the process of spectrum acquisition using the portable near-infrared spectrometer, the room temperature must be maintained at 25°C. When collecting the spectrum of each pea sample, the temperature of the portable near-infrared spectrometer must be kept consistent, otherwise the information carried by the spectrum will change, causing baseline drift. Therefore, when using the portable near-infrared spectrometer, it is necessary to first preheat it at room temperature for 1h after turning it on, and then measure it after the temperature of the portable near-infrared spectrometer reaches a constant. The pea sample is placed in a sample cup (50mm high; 51mm in diameter), and gently shaken to ensure that the pea seeds are evenly distributed. Each pea sample is scanned 5 times, and the sample cup is rotated a certain angle each time it is scanned. Repeat the sample loading and scanning three times to take the average value. Due to the influence of sample heterogeneity and distribution heterogeneity, the spectra of different parts of a sample are different. Therefore, it is necessary to mix and scan multiple times to obtain the average spectrum as the original spectrum to reduce the error.

[0074] Furthermore, the pea near-infrared component detection model described in this embodiment is established using TheUnscrambler X 10.3 software (CAMO, Norway) and MATLAB R2021a software (MathWorks, USA).

[0075] Furthermore, in step 4 described in this embodiment, MATLAB R2021a software is used to write principal component analysis programming (PCA) combined with the Mahalanobis distance program to screen out abnormal samples from the near-infrared reflectance spectra and basic component content values ​​of all pea samples, and the near-infrared spectral data of the remaining qualified sample sets are preprocessed by the Unscrambler X 10.3 software to establish an accurate and robust pea near-infrared component detection model for the basic component content of peas.

[0076] Preferably, the Mahalanobis distance is a generalized square distance, based on the theory of multivariate normal distribution, effectively considering the three parameters of mean, variance and covariance, and is a comprehensive indicator that can fully describe the overall multivariate structure; the Mahalanobis distance reflects the difference between samples by calculating the distance between different sample data and the center value of the sample set. Principal Component Analysis (PCA) is a mathematical method for dimensionality reduction. It recombines a set of multiple variables into a new set of independent comprehensive variables through orthogonal transformation to reduce the number of variables, and selects principal components (PCs) with a larger contribution rate to represent the information of the original variables.

[0077] Furthermore, in step 4 of this embodiment, the Kennard-stone algorithm (3:1) is used to divide the calibration set and the validation set samples in the process of establishing the pea near-infrared component detection model. The Kennard-stone algorithm is to consider all samples as candidate samples of the calibration set, first select a pair of samples with the farthest Euclidean distance, and then calculate the Euclidean distance of each remaining sample to each known sample in the validation set, find the two samples farthest and closest to the selected sample, and select these two samples into the calibration set, and repeat this process until the required number of samples is reached.

[0078] Furthermore, the number of pea samples selected in this embodiment is greater than or equal to 90.

[0079] Preferably, the original spectrum of the pea sample is as follows Figure 3 As shown in the figure, the horizontal axis is the wavelength (900-1700nm) and the vertical axis is the absorbance. As shown in the figure, the absorption peak intensities of the pea sample spectra obtained by the portable near-infrared spectroscopy system are different, and the trend performance is roughly similar and smooth, and basically does not contain noise information. However, the original spectrum absorbance range is too large, which is not conducive to the establishment of a quantitative analysis model. Therefore, a single preprocessing method or a composite preprocessing method is required to optimize the original spectrum data. The result shows that the pea near-infrared spectrum data of this embodiment meets the requirements for establishing a near-infrared quantitative analysis model.

[0080] Preferably, the commonly used spectral preprocessing methods mainly include smoothing (Normalization), benchmarking (Baseline), standard normal variable transformation (SNV), detrending (Detrend), multivariate scatter correction (MSC), first-order derivative (1stDerivative) and second-order derivative (2ndDerivative). The present invention selects two, three or even four of the above seven single preprocessing methods in combination for composite spectral preprocessing, and finally selects the best spectral preprocessing method.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A near infrared system for detecting the content of pea ingredients and grading their quality, characterized by: It includes a detection module, an analysis and judgment module and a storage module, and the detection module, the analysis and judgment module and the storage module can all be arranged in a portable near-infrared spectrometer; The detection module is used to perform near-infrared detection on peas and obtain the near-infrared reflectance spectrum of peas; The analysis and judgment module can analyze and judge the pea near-infrared reflectance spectrum obtained by the detection module through the pea near-infrared component detection model to obtain the specific pea component content and grade the component content, and finally output the content of various pea components and the corresponding grades; The storage module is used to store the results obtained by the analysis and judgment module and output and display them through the display screen of the portable near-infrared spectrometer.

2. A near infrared system for detecting the content of pea components and grading the quality according to claim 1, characterized in that: The ingredients in step 2 include protein, starch, water, crude fiber and crude fat; The first-grade protein content is greater than or equal to 25%, the second-grade content is 20-25%, and the third-grade content is less than or equal to 20%; The starch first grade content is greater than or equal to 55%, the second grade content is 45-55%, and the third grade content is less than or equal to 45%; The first level of moisture content is greater than or equal to 13%, the second level is 10-13%, and the third level is less than or equal to 10%; The crude fiber has a first-grade content of greater than or equal to 7%, a second-grade content of 5-7%, and a third-grade content of less than or equal to 5%; The first grade crude fat content is less than or equal to 5%, the second grade content is 5-9%, and the third grade content is greater than or equal to 9%.

3. A near infrared system for detecting the content of pea components and grading the quality according to claim 1, characterized in that: The method for constructing the pea near-infrared component detection model comprises the following steps: Step 1: Select peas of different varieties from different origins as samples and screen out impurities and damaged peas before labeling the samples; Step 2: All pea samples are tested using a portable near-infrared spectrometer to obtain near-infrared reflectance spectra of the corresponding pea samples; Step 3: using chemical methods to determine the basic component contents of all pea samples obtained from step 2 with near infrared reflectance spectra; Step 4: Use the partial least squares regression correction method to establish a pea near-infrared component detection model based on the near-infrared reflectance spectrum of the pea sample and the basic component content of the pea sample.

4. A near infrared system for detecting the content of pea components and grading the quality according to claim 3, characterized in that: The pea samples after screening in the step 1 are placed in a refrigerator and stored at a temperature of 4°C.

5. A near infrared system for detecting the content of pea components and grading the quality according to claim 3, characterized in that: Before the portable near-infrared spectrometer obtains the near-infrared reflectance spectrum, the portable near-infrared spectrometer and the pea sample are placed in the same environment for 24 hours to keep the pea sample environment consistent with the operating environment of the portable near-infrared spectrometer.

6. A near infrared system for detecting the content of pea components and grading the quality according to claim 3, characterized in that: In the step 1, the width, length and thickness of the pea sample after screening are repeatedly analyzed five times, and the average value is taken; the 100-grain weight of the pea sample after screening is weighed by an analytical balance, and the weighing is repeated three times to take the average value.

7. A near infrared system for detecting the content of pea components and grading the quality according to claim 3, characterized in that: The basic component contents of the pea samples include starch, crude protein, crude fiber, moisture and crude fat. The chemical analysis of the basic components of each pea sample is repeated three times, and the results are averaged for data analysis; the starch content of the pea samples is determined by a fully automatic starch detector; the crude protein content of the pea samples is determined by a Kjeldahl nitrogen determination method and a 2300 nitrogen analyzer with a conversion coefficient of 5.46; the moisture content of the pea samples is detected by a drying method; the crude fat content of the pea samples is determined by an automatic Soxhlet total fat analyzer; and the crude fiber of the pea samples is determined by a semi-automatic fiber analyzer.

8. A near infrared system for detecting the content of pea components and grading the quality according to claim 3, characterized in that: The spectral range of the portable near-infrared spectrometer is 908-1676nm, the variable is 90, the sampling interval is 6nm, the light source is a 10W halogen lamp, and the spectral integration time is 12.7ms.

9. A near infrared system for detecting the content of pea components and grading the quality according to claim 3, characterized in that: In the step 4, the Kennard-stone algorithm is used to divide the calibration set and the validation set samples during the establishment of the pea near-infrared component detection model.

10. A near infrared system for detecting the content of pea components and grading the quality according to claim 3, characterized in that: The number of pea samples selected is greater than or equal to 90.

Images