Sorghum variety discrimination method and system based on decision fusion and near infrared spectrum

Abstract

The invention discloses a sorghum variety discrimination method and system based on decision fusion and near infrared spectroscopy. The method comprises the following steps: collecting and processing samples of sorghum grains; collecting a near infrared spectrum of each sample, and constructing a calibration set; determining the tannin content and amylopectin content of each sample; constructing a first quantitative detection model between the near infrared spectrum of the sample and the tannin content, and constructing a second quantitative detection model between the near infrared spectrum of the sample and the amylopectin content; constructing a qualitative discrimination model between the near infrared spectrum of the sample and the category affiliation; respectively predicting the correction set to obtain three groups of predicted values, and constructing a decision fusion model by taking the three groups of predicted values as independent variables and the category labels of the samples as dependent variables; respectively predicting the tannin content and amylopectin content of an unknown sorghum sample, and predicting the category attribution of the unknown sorghum sample by adopting a qualitative discrimination model to obtain a final category prediction value; the method has the advantages of high precision and strong robustness.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of sorghum variety classification, and in particular to a method and system for classifying sorghum varieties based on decision fusion and near-infrared spectroscopy. BACKGROUND

[0002] Sorghum is one of the important food crops in China. Sorghum is rich in various nutrients and is mainly used as a raw material for brewing wine in China. Different wine enterprises use different sorghum varieties for brewing, and in the raw material procurement process, in order to ensure the quality of the processed wine, the enterprises give higher procurement prices for their special sorghum varieties than for other varieties. Traditional sorghum variety identification methods such as molecular identification and field identification are complex and time-consuming, and cannot meet the variety determination needs of wine enterprises in the procurement and storage process. Near-infrared spectroscopy technology, as a mature rapid detection technology, is very suitable for rapid identification of sorghum varieties. For example, Chinese patent publication CN103743705A discloses a rapid detection method for false sorghum and its similar species, determines the optimal principal component number as 9, after feature extraction, takes the 9 principal components as input variables of the neural network, establishes a three-layer BP neural network, and calculates the category to which each subcategory belongs. This patent application and other similar technical solutions that use spectroscopy to identify crop varieties are representative. They all use the category label as the output layer, use different algorithms to obtain the spectrum or some characteristic variables of the spectrum as the input layer, and then build a relationship model between the two.

[0003] However, using the category label as the output layer has certain limitations. The category label is usually a limited number of discrete values. Using it as the output for modeling can lead to insufficient model interpretability and the model is extremely dependent on the calibration sample. When the calibration sample is not representative, the model is not robust. Assuming that the calibration sample not only has differences in varieties, but also has differences other than varieties, if only the category label is used as the output value, then no matter what improved algorithm is used, it cannot guarantee a robust and reliable prediction result. To give a specific example, assume that Guojiaohong No. 1 from Luzhou and Qingqiao Yangsuo sorghum from Hainan are used as calibration samples for binary classification. If the category label (0 or 1) is used as the output value, the actual model built can be interpreted as either a variety discrimination model or a production area discrimination model, because the labels for these two types of models are the same. When the actual test sample and the calibration sample differ, for example, the actual test sample is Guojiaohong No. 1 from Hainan, the model may make a wrong prediction based on the production area rather than the variety of the object. Therefore, near-infrared qualitative discrimination analysis usually requires hundreds or thousands of samples to increase the representativeness of the calibration set as much as possible, but in actual application, collecting so many samples requires a high time cost and labor cost. SUMMARY

[0004] The technical problems to be solved by the present application are that the existing technical method for identifying sorghum varieties has poor interpretability, low precision and poor robustness, and the existing spectral detection technology has great dependence on sample representativeness.

[0005] The present application solves the above technical problems by the following technical means: a method for identifying sorghum varieties based on decision fusion and near-infrared spectrum, comprising: S1. Collecting and processing samples of sorghum grains; S2. Collecting near-infrared spectrum of each sample to construct a calibration set; S3. Determining the tannin content and amylopectin content of each sample; S4. Constructing a first quantitative detection model between the near-infrared spectrum of the sample and the tannin content, and constructing a second quantitative detection model between the near-infrared spectrum of the sample and the amylopectin content; S5. Constructing a qualitative discrimination model between the near-infrared spectrum of the sample and the category attribution; S6. Using the first quantitative detection model, the second quantitative detection model and the qualitative discrimination model to respectively predict the calibration set to obtain three groups of predicted values, taking the three groups of predicted values as independent variables and the category label of the sample as dependent variables to construct a decision fusion model; S7. For an unknown sorghum sample, collect the spectrum in the same way as in S2, predict its tannin content and amylopectin content using the first quantitative detection model and the second quantitative detection model in S4, predict its category attribution using the qualitative discrimination model in S5, and then further predict the above three groups of predicted values using the decision fusion model in S6 to obtain the final category prediction value. When the final category prediction value is ≤0.5, the sample belongs to other varieties, and when the final category prediction value is >0.5, the sample belongs to the target variety.

[0006] Further, the S1 comprises: Collect more than 60 samples of complete mature sorghum grains with known variety attribution to form a data set, which includes a single variety sample or multiple variety samples or multiple variety mixed samples that a user is interested in. These samples are regarded as target variety samples, and there are more than 30 target variety samples. The samples other than the target variety samples in the data set are regarded as other variety samples, and there are more than 30 other variety samples. Use a flour mill to process the sorghum grains of each target variety sample and other variety sample into powder. Assign a category label 1 to the target variety sample and a category label 0 to the other variety sample.

[0007] Further, the S2 comprises: The near-infrared spectrum of each sample was collected on a near-infrared spectrometer in a diffuse reflection mode, and each sample was collected twice at different time periods, the near-infrared spectrum of the first collected sample was used as a calibration set for modeling, and the near-infrared spectrum of the second repeatedly collected sample was used as a validation set for verifying the prediction performance of the decision fusion model.

[0008] Further, the S3 comprises: The tannin content and amylopectin content of each sorghum sample were determined by a wet chemical method.

[0009] Further, the determination method of the amylopectin content comprises: The standard solution preparation and absorbance detection were performed according to the standard NY / T 55-1987 of the Ministry of Agriculture and Rural Affairs of the People's Republic of China, the absorbance of the prepared standard solution at 640 nm and 532 nm was read by a spectrophotometer, the absorbance read at 640 nm constituted a vector , and the absorbance read at 532 nm constituted a vector , the vector and the vector constituted a matrix , the matrix was used as an independent variable, the amylopectin concentration was used as a dependent variable, and a partial least squares method was used to construct an amylopectin concentration regression equation; the absorbance of the sample to be measured at 640 nm and 532 nm was read, and the amylopectin concentration regression equation was substituted to obtain the amylopectin concentration , and the amylopectin content was obtained by using the formula , wherein the mass of the sample to be measured, and the moisture percentage of the sample to be measured.

[0010] Further, the construction process of the amylopectin concentration regression equation is: S321, the matrix is standardized by using the formula , the standardized , represents the mean, and the standard deviation; the amylopectin concentration is standardized by using the following formula (1) , wherein is the standardized ; The first weight vector of is obtained by solving the formula , and the value of the first weight vector is obtained by using the formula , wherein, the largest eigenvalue of the matrix is calculated as where represents the contribution weight of to the construction of the first score vector represents the contribution weight of to the construction of the first score vector ; the first score vector and the second score vector are calculated, wherein represents the first weight vector of ; the first loading vector of and the first loading vector of are calculated ;

[0011] the first residual matrix of and the first residual matrix of are calculated ; ; S322, the residual matrix and obtained in the last iteration are taken as the new and , and all steps of S321 are repeated to obtain the second weight vector of 、 the third score vector of the fourth score vector of the second loading vector of the second loading vector of the second residual matrix of and the second residual matrix of is calculated as wherein, represents the contribution weight of to the construction of , represents the contribution weight of to the construction of ; S323, regression is performed on , and the formula is:​​ (2) wherein, is the residual of, and are regression coefficients of, , , ; S324, the and are expressed by : (3) (4) According to formula (1) and (2), , combined with formula (3) and (4), after merging the same term, the final formula is obtained as follows:

[0012] wherein, is the intercept, and are regression coefficients of, , and the specific formula is:

[0013]

[0014] 。

[0015] Further, the first quantitative detection model and the second quantitative detection model are both modeled by using the partial least squares method, and the modeling parameters of the first quantitative detection model are as follows: first, the sample spectrum is processed by using the Savitsky-Golay method 1st order derivative 17-point smoothing, then the near-infrared spectrum at the range of 928-1008nm, 1174-1328nm and 1550-1617nm is intercepted for modeling, and the PLS factor number during modeling is 9; The modeling parameters of the second quantitative detection model are: first derivative 17-point smoothing processing of the sample spectrum by the Savitsky-Golay method, then near-infrared spectra at 938 nm, 960 nm, 1034-1037 nm, 1043-1052 nm, 1062-1065 nm, 1109-1118 nm, 1124-1128 nm, 1156-1161 nm, 1214-1223 nm, 1244 nm, 1250 nm, 1259 nm, 1265 nm, 1277 nm, 1315-1325 nm, 1378-1381 nm, 1430 nm, 1445 nm, 1456 nm, 1502 nm, 1508 nm, 1513 nm, 1653 nm and 1686 nm are intercepted for modeling, and the PLS factor number is 10 during modeling.

[0016] Further, the qualitative discrimination model is modeled by the partial least squares method, and the modeling parameters of the qualitative discrimination model are: The near-infrared spectra at 928 nm, 996 nm, 1087 nm, 1223 nm, 1322 nm, 1345 nm, 1669 nm and 1680 nm are intercepted for modeling, and the PLS factor number is 8 during modeling.

[0017] Further, the decision fusion model is modeled by the discriminant partial least squares method, and the modeling parameters of the decision fusion model are that the PLS factor number is 3.

[0018] The application also provides a discrimination system for sorghum varieties based on decision fusion and near-infrared spectroscopy, comprising: A sample collection module is used for sample collection and processing of sorghum kernels. A spectrum collection module is used for collecting near-infrared spectra of each sample to construct a calibration set. A chemical reference value acquisition module is used for determining the tannin content and amylopectin content of each sample. A quantitative model construction module is used for constructing a first quantitative detection model between the near-infrared spectrum and the tannin content of the sample, and constructing a second quantitative detection model between the near-infrared spectrum and the amylopectin content of the sample. A qualitative discrimination model construction module is used for constructing a qualitative discrimination model between the near-infrared spectrum and the category attribution of the sample. A decision fusion model construction module is used for predicting the calibration set by using the first quantitative detection model, the second quantitative detection model and the qualitative discrimination model respectively to obtain three groups of predicted values, taking the three groups of predicted values as independent variables, and taking the category label of the sample as a dependent variable to construct a decision fusion model. The result prediction module is configured to, for an unknown sorghum sample, collect a spectrum in the same manner as the spectrum collection module, use the first quantitative detection model and the second quantitative detection model in the quantitative model construction module to respectively predict the tannin content and the amylopectin content of the sample, use the qualitative discrimination model in the qualitative discrimination model construction module to predict the category attribution of the sample, and then use the decision fusion model in the decision fusion model construction module to further predict the three groups of prediction values to obtain a final category prediction value. When the final category prediction value is less than or equal to 0.5, the sample belongs to other varieties, and when the final category prediction value is greater than 0.5, the sample belongs to the target variety.

[0019] Further, the sample collection module is further configured to: collect 60 or more complete mature sorghum grain samples of which the varieties are known to constitute a data set, wherein the data set includes a single variety sample or multiple variety samples or multiple variety mixed samples that are of interest to a user, and these samples are regarded as target variety samples, and the target variety samples are 30 or more; samples other than the target variety samples in the data set are regarded as other variety samples, and the other variety samples are 30 or more; and a flour mill is used to process the sorghum grains of each target variety sample and other variety sample into powder; the target variety samples are assigned a category label 1, and the other variety samples are assigned a category label 0.

[0020] Further, the spectrum collection module is further configured to: collect the near-infrared spectrum of each sample in a diffuse reflection manner on a near-infrared spectrometer, and collect the near-infrared spectrum of each sample twice in different time periods, and use the first collected near-infrared spectrum of the sample as a calibration set for modeling, and use the second repeatedly collected near-infrared spectrum of the sample as a validation set for validating the prediction performance of the decision fusion model.

[0021] Further, the chemical reference value acquisition module is further configured to: use a wet chemical method to determine the tannin content and the amylopectin content of each sorghum sample.

[0022] Still further, the method for determining the amylopectin content comprises: use the standard solution preparation and absorbance detection of the Chinese Ministry of Agriculture and Rural Affairs standard NY / T 55-1987, use a spectrophotometer to read the absorbance of the prepared standard solution at 640 nm and 532 nm, use the absorbance read at 640 nm to form a vector and use the absorbance read at 532 nm to form a vector , the vector and the vector form a matrix , and use the matrix as an independent variable, and the amylopectin concentration For the dependent variable, a regression equation of amylopectin concentration is constructed by using partial least squares method; the absorbance of the sample to be measured at 640 nm and 532 nm is read, and the amylopectin concentration is obtained by substituting the absorbance into the regression equation of amylopectin concentration , the amylopectin content is obtained by using the formula wherein, the mass of the sample to be measured, the moisture percentage of the sample to be measured.

[0023] Further, the construction process of the regression equation of amylopectin concentration is as follows: S321, the matrix is standardized by using the formula , is the standardized , represents the mean, represents the standard deviation; the amylopectin concentration is standardized by using the formula (1) wherein, is the standardized ; the first weight vector of is obtained by solving the formula , wherein is the maximum eigenvalue of the matrix , and the calculated is represented as wherein represents the contribution weight of to the construction of the first score vector , represents the contribution weight of to the construction of the first score vector ; the first score vector and the second score vector are calculated, wherein represents the first weight vector of ; the first loading vector of and the first loading vector of

[0024] the first residual matrix of and the first residual matrix of are calculated; ​S322, the residual matrix obtained in the last iteration and as the new and , repeat all steps of S321 to obtain the second weight vector 、 the third score vector , the fourth score vector , the second load vector , the second load vector , the second residual matrix and the second residual matrix , the calculated is expressed as , wherein represents the contribution weight of to the construction of , and represents the contribution weight of to the construction of ; S323, regression is performed on , and the formula is: (2) wherein, is the residual of , which is the error term left after fitting by formula (2), and are regression coefficients of , which can be obtained by least squares, and the formula is: , ; S324, the and are expressed by : (3) (4) According to formulas (1) and (2), , combined with formulas (3) and (4), after combining like terms, the final formula is obtained as follows:

[0025] wherein, is the intercept, and are the regression coefficient of the formula is:

[0026]

[0027] 。

[0028] Further, the first quantitative detection model and the second quantitative detection model are both modeled by using a partial least squares method, modeling parameters of the first quantitative detection model are as follows: first, the sample spectrum is processed by using a Savitsky-Golay method 1st derivative 17-point smoothing, then, near-infrared spectra at 928-1008 nm, 1174-1328 nm and 1550-1617 nm are intercepted for modeling, and the PLS factor number is 9 during modeling; modeling parameters of the second quantitative detection model are as follows: first, the sample spectrum is processed by using a Savitsky-Golay method 1st derivative 17-point smoothing, then, near-infrared spectra at 938 nm, 960 nm, 1034-1037 nm, 1043-1052 nm, 1062-1065 nm, 1109-1118 nm, 1124-1128 nm, 1156-1161 nm, 1214-1223 nm, 1244 nm, 1250 nm, 1259 nm, 1265 nm, 1277 nm, 1315-1325 nm, 1378-1381 nm, 1430 nm, 1445 nm, 1456 nm, 1502 nm, 1508 nm, 1513 nm, 1653 nm and 1686 nm are intercepted for modeling, and the PLS factor number is 10 during modeling.

[0029] Further, the qualitative discrimination model is modeled by using a partial least squares method, and modeling parameters of the qualitative discrimination model are as follows: near-infrared spectra at 928 nm, 996 nm, 1087 nm, 1223 nm, 1322 nm, 1345 nm, 1669 nm and 1680 nm are intercepted for modeling, and the PLS factor number is 8 during modeling.

[0030] Further, the decision fusion model is modeled by using a discriminant partial least squares method, and modeling parameters of the decision fusion model are as follows: the PLS factor number is 3.

[0031] The present application has the following advantages: (1) The present application proposes a near-infrared discrimination method based on decision fusion, on the basis of the traditional classification model of sorghum varieties based on category labels, a plurality of near-infrared quantitative detection models of sorghum components are added. Then the basic sorghum variety classification model (i.e. qualitative discrimination model), the newly added near-infrared quantitative detection model of sorghum components is used to predict the sample, and the classification prediction value and the component prediction value are used as the input to build a new variety classification model, i.e. a decision layer fusion model, to re-determine the category attribution of sorghum varieties. Its beneficial effect lies in that the method considers not only the traditional category attribution of sorghum varieties, but also the actual correlation between sorghum varieties and components. By constructing a decision layer fusion model, the component content (prediction value) of the sorghum sample is used to correct the initial variety classification result. Since the final result is based on the joint calculation of the qualitative discrimination model and the quantitative discrimination model, this correction method can enhance the model's interpretability while improving the discrimination effect and improving the discrimination precision. In near-infrared quantitative analysis, only a few tens of samples are needed to build a model, and the reason is that the output value of the quantitative model is not a discrete category label, but a continuous component value. The model can self-correct according to the difference between the several component values, reduce the dependence of the model on the representativeness of the sample in the case of few samples, and improve the prediction robustness.

[0032] (2) The present application takes the absorbance of 640nm and 532nm as independent variables X 1 and X 2, and takes the amylopectin concentration as dependent variable Y , and uses partial least squares method (PLS factor number is 2) to construct the amylopectin concentration regression equation, the main reason is that: when performing color reaction, absorption peaks in the range of 600-680nm and 520-555nm will be generated, which are the characteristic absorption peaks of linear starch-iodine complex and branched starch-iodine complex generated by reaction. Among them, 640nm is the maximum absorption wavelength of linear starch-iodine complex, and 532nm is the maximum absorption wavelength of branched starch-iodine complex, and the regression effect is theoretically best, so these two absorbance values are selected.

[0033] (3) The wave band range, pretreatment and PLS factor number in S4 and S5 of the present application are obtained by combination of a plurality of wave band selection algorithms and pretreatment algorithms, and the model is optimized and screened with the lowest cross-validation root mean square error as the model performance evaluation standard. Under the selection of related parameters, the model has higher precision and better performance. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 The flow chart of the discrimination method of sorghum varieties based on decision fusion and near-infrared spectrum disclosed in embodiment 1 of the present application; Figure 2The prediction value scatter plot of the discriminant model constructed in S5 and the decision fusion model constructed in S6 of the discrimination method of sorghum varieties based on decision fusion and near-infrared spectroscopy disclosed in Embodiment 1 of the present application for predicting the calibration set samples; Figure 3 The prediction value scatter plot of the discriminant model constructed in S5 and the decision fusion model constructed in S6 of the discrimination method of sorghum varieties based on decision fusion and near-infrared spectroscopy disclosed in Embodiment 1 of the present application for predicting the calibration set samples; DETAILED DESCRIPTION

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in a clear and complete manner with reference to the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0036] Embodiment 1 As shown in the following table, Embodiment 1 of the present application provides a discrimination method of sorghum varieties based on decision fusion and near-infrared spectroscopy, which comprises the following steps: Figure 1 S1. Sample collection and processing of sorghum grains: In this embodiment, in order to verify the method proposed in the present application, a set of data is designed, containing 81 grain samples of sorghum varieties, which are composed of 26 sorghum varieties from different regions. The samples of varieties belonging to Gujiao Hong No. 1 and Qingke Yang (or their mixed samples) are taken as a class, regarded as target variety samples, a total of 34 samples. The samples of sorghum varieties of the remaining varieties and origins are taken as another class, regarded as other variety samples, a total of 47 samples. Remove imperfect grains from each sample, and use a cyclone mill for milling treatment. The target variety samples are assigned a class label 1, and the other variety samples are assigned a label 0. S2. Spectral acquisition: Use a portable near-infrared spectrometer to carry out spectral acquisition. When collecting, place the sample in a sample cup with a transparent quartz glass at the bottom, place it above the instrument detection window, and use the diffuse reflection method to collect the spectrum, with a spectral range of 928-1685 nm and a resolution of 3 nm. Each sample is collected twice at different time intervals. The first collected sorghum powder spectrum is used as the calibration set for modeling, and the second repeated spectrum is used as the validation set for verifying the model prediction performance.

[0037] S3. Chemical reference value acquisition: The tannin content and amylopectin content of each sorghum sample are determined by wet chemical method.

[0038]

[0039] ​S31. The wet chemical detection method of tannin content refers to the Chinese national standard method GB / T 15686-2008.

[0040] S32. The wet chemical detection method of amylopectin content is improved based on the Chinese Ministry of Agriculture and Rural Affairs standard "NY / T 55-1987 Determination of amylose in rice, corn and millet grains". 1) Preparation of standard solution: Take 0.05 g of dried potato amylose pure product and place it in a 10 ml centrifuge tube. Wet the sample with 0.5 ml of anhydrous ethanol, then add 4.5 ml of 1 mol / L sodium hydroxide solution. Disperse in a boiling water bath for 10 minutes, then dilute to 10 ml with water to obtain a 5 mg / ml amylose standard solution. Take 0.05 g of dried sorghum amylopectin pure product and place it in a 10 ml centrifuge tube. Wet the sample with 0.5 ml of anhydrous ethanol, then add 4.5 ml of 1 mol / L sodium hydroxide solution. Disperse in a boiling water bath for 10 minutes, then dilute to 10 ml with water to obtain a 5 mg / ml amylopectin standard solution.

[0041] 2) Standard sample detection: Take 6 50 ml volumetric flasks and add 0, 25, 50, 100, 150, and 200 μl of 5 mg / ml amylose standard solution, respectively. Then add 500 μl, 475 μl, 450 μl, 400 μl, 350 μl, and 300 μl of 5 mg / ml amylopectin standard solution, respectively, to a total volume of 500 μl. Take another 100 ml volumetric flask and add 5 ml of 0.09 mol / L sodium hydroxide solution as a blank. Then add about 25 ml of water, 0.5 ml of 1 mol / L acetic acid, and 0.5 ml of iodine reagent (iodine reagent is prepared by dissolving 2 g of iodine and 20 g of potassium iodide in distilled water and diluting to 100 ml, then taking 10 ml and diluting to 100 ml) to the volumetric flasks except the blank, and dilute to 50 ml with water. Add 50 ml of water, 1 ml of 1 mol / L acetic acid, and 1 ml of iodine reagent to the blank volumetric flask, and dilute to 100 ml with water. Develop the color for 10 minutes, use the blank as a control, and read the absorbance at 640 nm and 532 nm using a spectrophotometer.

[0042] It should be noted that when the color reaction is carried out, the absorption peaks in the range of 600-680 nm and 520-555 nm are generated, which are the characteristic absorption peaks of the amylose-iodine complex and the amylopectin-iodine complex generated by the reaction respectively. Among them, 640 nm is the maximum absorption wavelength of the amylose-iodine complex, 532 nm is the maximum absorption wavelength of the amylopectin-iodine complex, and the regression effect is theoretically the best, so these two absorbance values are selected. The standard solution configuration and the solution ratio of the absorbance detection process of the standard solution configuration and the absorbance detection process of the standard solution configuration disclosed in the standard "NY / T 55-1987 Determination of amylose in rice, maize and millet grains" of the Ministry of Agriculture and Rural Affairs of the People's Republic of China are adjusted, and then the absorbance at two spectral points is read and the mixed calibration curve is drawn, and "NY / T 55-1987 Determination of amylose in rice, maize and millet grains" is read at a single spectral point and the calibration curve is drawn. In practical application, the standard solution configuration and the absorbance detection method disclosed in the standard "NY / T 55-1987 Determination of amylose in rice, maize and millet grains" of the Ministry of Agriculture and Rural Affairs of the People's Republic of China can be directly used, and the ratio is not improved. The specific selection is made according to the actual needs. The following will introduce the part of the amylopectin content determination process which is greatly improved compared with the Chinese national standard in detail.

[0043] 3) Mixed calibration curve drawing: the absorbance of the 6 50ml volumetric flasks at 640nm constitutes a vector and the absorbance of the 6 50ml volumetric flasks at 532nm constitutes a vector , the vector and the vector constitute a matrix , with the matrix as the independent variable and the amylopectin concentration as the dependent variable, the partial least squares method (PLS factor number is 2) is used to construct the amylopectin concentration regression equation, and the specific steps are as follows: S321, extract the first PLS component: (1) Calculate the first weight vector of : First, standardize using the formula , and the standardized is , wherein represents the mean value, represents the standard deviation; the amylopectin concentration is standardized by the following formula (1) , wherein is the standardized ; Solving formula Get the first weight vector The value of, wherein, is the largest eigenvalue of the matrix , Need to be unitized (module length is 1). After calculation Can be expressed as Where Represent The contribution weight of the first score vector , Represent The contribution weight of the first score vector .

[0044] (2) Calculate the first PLS component score of the first score vector And the second score vector ( ): Let 1, calculate ， Where Indicates the first weight vector of , which is paired with , and the calculation method is the same, except that The And Of the calculation process are exchanged, which will not be repeated here.

[0045] (3) Calculate the first loading vector Of And the first loading vector Of :

[0046]

[0047] (4) Calculate the residual matrix Use the extracted first component to explain And , and then calculate the remaining unexplained part (residual), the formula is:

[0048]

[0049] Where, Indicates the first residual matrix of , Indicates the first residual matrix of .

[0050] S322. Extract the second PLS component, including the following steps: The residual matrix obtained from the previous iteration and As a new and Repeat all steps in S321 to extract the second component pair. The calculation yields... The second weight vector 、 The third score vector , The fourth score vector , The second load vector , The second load vector , The second residual matrix and The second residual matrix After calculation Represented as ,in, represent For building Contribution weight, represent For building The contribution weight.

[0051] S323. Establish the regression equation: right The regression formula is as follows: (2) for The residual is the error term remaining after fitting the equation (2). and All The regression coefficients can be obtained using the least squares method, and the formula is as follows: , .

[0052] S324. Convert the standardized equations back to the original dimensions: Will and use express: (3) (4) According to formulas (1) and (2), it can be seen that 。

[0053] By further combining formulas (3) and (4) and merging like terms, we can finally obtain the following formula:

[0054] in It is the intercept. and yes The regression coefficient is given by the following formula:

[0055]

[0056] 。

[0057] 4) Sample determination: First, the moisture content of the sample was determined using the drying and weighing method. Next, the equivalent of 0.05g of crude starch was weighed into a 10ml centrifuge tube, and 0.5ml of anhydrous ethanol was added to thoroughly moisten the sample. Then, 4.5ml of 1mol / L sodium hydroxide solution was added, the tube was capped, and the mixture was shaken for 10 seconds. Afterward, it was dispersed in a boiling water bath for 10 minutes, rapidly cooled, and then diluted to 10ml with water. Finally, 0.5ml of the sample was pipetteed into a 50ml volumetric flask, 25ml of water was added, followed by 0.5ml of 1mol / L acetic acid solution and 0.5ml of iodine reagent. The volume was then diluted to 50ml with water, and the sample was developed for 10 minutes. The absorbance was read at 640nm and 532nm using a spectrophotometer.

[0058] 5) Result Calculation: For the absorbance of each sample at 640 nm and 532 nm, the amylopectin concentration of the sample is predicted according to the curve in 3). (%, as a percentage of total sample weight), and then calculate the amylopectin content G0 (%, as a percentage of sample dry weight) using the following formula: (2) in The mass of the sample used to determine the amylopectin content, The moisture percentage of the sample used to calculate the amylopectin content.

[0059] The statistical results of the tannin and amylopectin content of sorghum determined by the above chemical methods are shown in Table 1. Both tannin and amylopectin contents have a relatively wide distribution, covering most types of sorghum used for brewing.

[0060] Table 1. Statistical results of tannin and amylopectin content in sorghum correction set.

[0061] S4. Quantitative Model Construction: A quantitative detection model between the spectral density and tannin content of each sorghum sample was constructed using partial least squares method. Similarly, a quantitative detection model between the spectral density and amylopectin content of each sorghum sample was constructed using partial least squares method.

[0062] Both sets of models were selected from different combinations of preprocessing methods and band selection algorithms. After optimization, the optimal modeling parameters were determined, as shown in Table 2.

[0063] Table 2 Modeling parameters and performance of the sorghum composition model

[0064] Table 2 shows the cross-validation determination coefficients (R²) for the two models. 2 cv The correlation coefficients (R²) of the two models were relatively high, at 0.854 and 0.882 respectively, and the root mean square error of cross-validation (RMSECV) were relatively low, at 0.151 and 1.080 respectively, indicating good modeling performance. When using the validation set samples for prediction, the prediction correlation coefficients (R²) of the two models were relatively high. p The values ​​were 0.889 and 0.867, respectively, and the root mean square error of prediction (RMSEP) were 0.187 and 1.689, respectively, indicating that the prediction performance was also good.

[0065] S5. Qualitative Discriminant Model Construction: Qualitative discriminant models were constructed for the spectral composition and category of each sorghum sample using the discriminant partial least squares (PLS) method. The modeling parameters were as follows: near-infrared spectra at 928 nm, 996 nm, 1087 nm, 1223 nm, 1322 nm, 1345 nm, 1669 nm, and 1680 nm were used for modeling, and the PLS factor was 8. When the spectral prediction value of a sample was ≤0.5, the sample belonged to the "Other Variety Samples" category; when the spectral prediction value was >0.5, the sample belonged to the "Target Variety Samples" category.

[0066] S6. Construction of Decision Fusion Model: Using the near-infrared spectra of the model prediction calibration set obtained in steps S4 and S5, three sets of predicted values ​​are obtained (predicted tannin content, predicted amylopectin content, and predicted category classification, respectively). These three sets of predicted values ​​are used as independent variables, and the category label of the sample is used as the dependent variable. A qualitative discriminant model is constructed between them using the discriminant partial least squares method. The PLS factor number of the model is 3. This model is the decision fusion model.

[0067] S7. Prediction: For unknown sorghum samples, the sorghum is processed into powder, and spectra are collected using the same method as in step S2. The quantitative model in step S4 is used to predict its tannin and amylopectin content, and the qualitative discrimination model in step S5 is used to predict its category. Then, the decision fusion model in step S6 is used to further predict the above three sets of predicted values ​​to obtain the final category prediction value. When the spectral prediction value of a sample is ≤0.5, the sample belongs to the "Other Variety Samples" category; when the spectral prediction value of a sample is >0.5, the sample belongs to the "Target Variety Samples" category.

[0068] To clearly illustrate the advantages of the proposed method compared to traditional near-infrared qualitative discrimination methods, the decision fusion model in S6 was used as the experimental group, and the qualitative model in S5 was used as the control group. The recognition performance of the two models on the calibration set and the validation sample set was statistically analyzed, as shown in Table 3.

[0069] Table 3 Comparison of modeling and predictive performance of sorghum variety discrimination models

[0070] Table 3 shows that the sorghum variety discrimination model optimized by conventional chemometrics methods before decision fusion had 4 misclassifications for the calibration samples, with an accuracy of 0.951 and an F1 score of 0.941. After further calculation by the decision fusion model in step S6, the number of misclassifications for the calibration samples decreased to 2, and the accuracy and F1 score improved to 0.975 and 0.971, respectively. In prediction, the control model had 5 misclassifications, with an accuracy of 0.938 and an F1 score of 0.921. After further calculation by the decision fusion model in step S6, the number of misclassifications for the predicted samples decreased to 4, and the accuracy and F1 score improved to 0.951 and 0.943, respectively. This indicates that the method of the present invention has a certain improvement in discrimination performance compared to the general near-infrared method for sorghum variety discrimination.

[0071] Figure 2 This is a scatter plot of the predicted values ​​of the experimental and control group model prediction calibration set samples. Figure 3 This is a scatter plot of the predicted values ​​for the model's validation set samples in the experimental and control groups. Figure 2 and Figure 3 As can be seen, after optimization by the method of this invention, the prediction results of the model tend to change towards the correct classification direction. A considerable number of samples belonging to other varieties tend to have lower predicted values, while samples belonging to the target variety tend to have higher predicted values. This allows samples near the classification threshold to be classified more correctly.

[0072] Through the above technical solutions, the rapid determination method for sorghum variety categories provided by the present invention has the advantages of objective and accurate detection results and rapid detection process, which can effectively improve the quality detection efficiency related to sorghum variety categories in the sorghum production, processing and storage process.

[0073] Example 2 Based on Example 1, Example 2 of the present invention also provides a sorghum variety discrimination system based on decision fusion and near-infrared spectroscopy, including: The sample collection module is used for the collection and processing of sorghum grain samples. The spectral acquisition module is used to acquire near-infrared spectra for each sample and construct a calibration set; The chemical reference value acquisition module is used to determine the tannin and amylopectin content of each sample; The quantitative model building module is used to build a first quantitative detection model between the near-infrared spectrum of a sample and the tannin content, and to build a second quantitative detection model between the near-infrared spectrum of a sample and the amylopectin content. The qualitative discrimination model building module is used to build a qualitative discrimination model between the near-infrared spectrum of a sample and its class classification. The decision fusion model construction module is used to predict the calibration set using the first quantitative detection model, the second quantitative detection model, and the qualitative discrimination model respectively, to obtain three sets of predicted values. These three sets of predicted values ​​are used as independent variables, and the class label of the sample is used as the dependent variable to construct the decision fusion model. The result prediction module is used to acquire spectra of unknown sorghum samples in the same way as the spectral acquisition module. The first and second quantitative detection models in the quantitative model construction module are used to predict the tannin content and amylopectin content, respectively. The qualitative discriminant model in the qualitative discriminant model construction module is used to predict the category. Then, the decision fusion model in the decision fusion model construction module is used to further predict the above three sets of predicted values ​​to obtain the final category prediction value. When the final category prediction value is ≤0.5, the sample belongs to other varieties, and when the final category prediction value is >0.5, the sample belongs to the target variety.

[0074] Specifically, the sample acquisition module is also used for: A dataset is constructed by collecting more than 60 complete and mature sorghum grain samples with known cultivar affiliations. The dataset includes samples of a single variety, multiple varieties, or a mixture of multiple varieties of interest to the user. These samples are considered as target variety samples, with more than 30 target variety samples. Samples other than target variety samples in the dataset are considered as other variety samples, with more than 30 other variety samples. The sorghum grains of each target variety sample and other variety samples are processed into powder using a mill. The target variety samples are assigned a category label of 1, and the other variety samples are assigned a category label of 0.

[0075] Specifically, the spectral acquisition module is also used for: Near-infrared spectra of each sample were acquired using a diffuse reflectance method on a near-infrared spectrometer. Each sample was acquired twice at different time periods. The near-infrared spectra of the first acquired sample were used as the calibration set for modeling, and the near-infrared spectra of the second acquired sample were used as the validation set to verify the predictive performance of the decision fusion model.

[0076] Specifically, the chemical reference value acquisition module is also used for: The tannin and amylopectin content of each sorghum sample was determined using a wet chemical method.

[0077] More specifically, the method for determining the amylopectin content includes: The preparation of standard solutions and absorbance detection were carried out according to the Chinese Ministry of Agriculture and Rural Affairs standard NY / T 55-1987. A spectrophotometer was used to read the absorbance of the prepared standard solutions at 640 nm and 532 nm. The absorbance read at 640 nm was used to construct a vector. The vector is formed by the absorbance read at 532 nm. ,vector sum vector Construct a matrix , with matrix amylopectin concentration is the independent variable. Using amylopectin as the dependent variable, a partial least squares regression equation for amylopectin concentration was constructed. The absorbance of the sample at 640 nm and 532 nm was read and substituted into the regression equation to obtain the amylopectin concentration. Using the formula Obtaining amylopectin content in, The quality of the sample to be tested, This represents the moisture percentage of the sample to be tested.

[0078] More specifically, the process of constructing the regression equation for amylopectin concentration is as follows: S321, Regarding the matrix Using formula Standardize, For standardization , Represents the mean. Representative standard deviation; for amylopectin concentration Standardization is performed using the following formula. (1) in, For standardization ; Solving the formula get First weight vector The value of , where, For matrix The largest eigenvalue, after calculation Represented as in represent To construct the first score vector Contribution weight, represent To construct the first score vector Contribution weight; Calculate the first score vector Second score vector ,in express The first weight vector; calculate First load vector and First load vector

[0079] calculate The first residual matrix and The first residual matrix ; S322. The residual matrix obtained from the previous iteration is... and As a new and Repeat all steps in S321 to obtain The second weight vector 、 The third score vector , The fourth score vector , The second load vector , The second load vector , The second residual matrix and The second residual matrix After calculation Represented as ,in, represent For building Contribution weight, represent For building Contribution weight; S323, to The regression formula is as follows: (2) in, for The residual, and All The regression coefficients, , ; S324, will and use express: (3) (4) According to formulas (1) and (2), Combining formulas (3) and (4), and after merging like terms, the final formula is as follows:

[0080] in, It is the intercept. and yes The regression coefficient is given by the following formula:

[0081]

[0082] 。

[0083] Specifically, both the first and second quantitative detection models are modeled using partial least squares (PLS). The modeling parameters for the first quantitative detection model are as follows: the sample spectrum is first smoothed using the first derivative of the Savitsky-Golay method with 17 points, and then the near-infrared spectrum in the ranges of 928-1008nm, 1174-1328nm, and 1550-1617nm is extracted for modeling, and the PLS factor number during modeling is 9. The modeling parameters of the second quantitative detection model are as follows: First, the sample spectrum is smoothed using the first derivative of the Savitsky-Golay method with 17 points. Then, the near-infrared spectra at 938nm, 960nm, 1034-1037nm, 1043-1052nm, 1062-1065nm, 1109-1118nm, 1124-1128nm, 1156-1161nm, 1214-1223nm, 1244nm, 1250nm, 1259nm, 1265nm, 1277nm, 1315-1325nm, 1378-1381nm, 1430nm, 1445nm, 1456nm, 1502nm, 1508nm, 1513nm, 1653nm, and 1686nm are used for modeling, and the PLS factor number during modeling is 10.

[0084] Specifically, the qualitative discrimination model is modeled using partial least squares, and the modeling parameters of the qualitative discrimination model are: Near-infrared spectra at 928nm, 996nm, 1087nm, 1223nm, 1322nm, 1345nm, 1669nm, and 1680nm were used for modeling, and the PLS factor was 8 during modeling.

[0085] Specifically, the decision fusion model is modeled using discriminant partial least squares (PLS) and the modeling parameter of the decision fusion model is 3 PLS factors.

[0086] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying sorghum varieties based on decision fusion and near-infrared spectroscopy, characterized in that, include: S1. Sampling and processing of sorghum grains; S2. Collect near-infrared spectra for each sample and construct a calibration set; S3. Determine the tannin and amylopectin content of each sample; S4. Construct a first quantitative detection model between the near-infrared spectrum of the sample and the tannin content, and construct a second quantitative detection model between the near-infrared spectrum of the sample and the amylopectin content. S5. Construct a qualitative discriminant model between the near-infrared spectrum of the sample and its class classification; S6. The calibration set is predicted using the first quantitative detection model, the second quantitative detection model, and the qualitative discrimination model respectively, and three sets of predicted values ​​are obtained. The three sets of predicted values ​​are used as independent variables and the class label of the sample is used as the dependent variable to construct a decision fusion model. S7. For unknown sorghum samples, the spectra are collected in the same way as in S2. The first quantitative detection model and the second quantitative detection module in S4 are used to predict the tannin content and amylopectin content, respectively. The qualitative discrimination model in S5 is used to predict the category. Then, the decision fusion model in S6 is used to further predict the above three sets of predicted values ​​to obtain the final category prediction value. When the final category prediction value is ≤0.5, the sample belongs to other varieties, and when the final category prediction value is >0.5, the sample belongs to the target variety.

2. The method for identifying sorghum varieties based on decision fusion and near-infrared spectroscopy according to claim 1, characterized in that, S1 includes: A dataset is constructed by collecting more than 60 complete and mature sorghum grain samples with known cultivar affiliations. The dataset includes samples of a single variety, multiple varieties, or a mixture of multiple varieties of interest to the user. These samples are considered as target variety samples, with more than 30 target variety samples. Samples other than target variety samples in the dataset are considered as other variety samples, with more than 30 other variety samples. The sorghum grains of each target variety sample and other variety samples are processed into powder using a mill. The target variety samples are assigned a category label of 1, and the other variety samples are assigned a category label of 0.

3. The method for identifying sorghum varieties based on decision fusion and near-infrared spectroscopy according to claim 1, characterized in that, S2 includes: Near-infrared spectra of each sample were acquired using a diffuse reflectance method on a near-infrared spectrometer. Each sample was acquired twice at different time periods. The near-infrared spectra of the first acquired sample were used as the calibration set for modeling, and the near-infrared spectra of the second acquired sample were used as the validation set to verify the predictive performance of the decision fusion model.

4. The method for identifying sorghum varieties based on decision fusion and near-infrared spectroscopy according to claim 1, characterized in that, S3 includes: The tannin and amylopectin content of each sorghum sample was determined using a wet chemical method.

5. The method for identifying sorghum varieties based on decision fusion and near-infrared spectroscopy according to claim 4, characterized in that, The method for determining the amylopectin content includes: The preparation of standard solutions and absorbance detection were carried out according to the Chinese Ministry of Agriculture and Rural Affairs standard NY / T 55-1987. A spectrophotometer was used to read the absorbance of the prepared standard solutions at 640 nm and 532 nm. The absorbance read at 640 nm was used to construct a vector. The vector is formed by the absorbance read at 532 nm. ,vector sum vector Construct a matrix , with matrix amylopectin concentration is the independent variable. Using amylopectin as the dependent variable, a partial least squares regression equation for amylopectin concentration was constructed. The absorbance of the sample at 640 nm and 532 nm was read and substituted into the regression equation to obtain the amylopectin concentration. Using the formula Obtaining amylopectin content in, The quality of the sample to be tested, This represents the moisture percentage of the sample to be tested.

6. The method for identifying sorghum varieties based on decision fusion and near-infrared spectroscopy according to claim 5, characterized in that, The process of constructing the regression equation for amylopectin concentration is as follows: S321, Regarding the matrix Using formula Standardize, For standardization , Represents the mean. Representative standard deviation; for amylopectin concentration Standardization is performed using the following formula. (1) in, For standardization ; Solving the formula get First weight vector The value of , where, For matrix The largest eigenvalue, after calculation Represented as in represent To construct the first score vector Contribution weight, represent To construct the first score vector Contribution weight; Calculate the first score vector Second score vector ,in express The first weight vector; calculate First load vector and First load vector calculate The first residual matrix and The first residual matrix ; S322. The residual matrix obtained from the previous iteration is... and As a new and Repeat all steps in S321 to obtain The second weight vector 、 The third score vector , The fourth score vector , The second load vector , The second load vector , The second residual matrix and The second residual matrix After calculation Represented as ,in, represent For building Contribution weight, represent For building Contribution weight; S323, to The regression formula is as follows: （2） in, for The residual, and All The regression coefficients, , ; S324, will and use express: （3） （4） According to formulas (1) and (2), Combining formulas (3) and (4), and after merging like terms, the final formula is as follows: in, It is the intercept. and yes The regression coefficient is given by the following formula: 。 7. The method for identifying sorghum varieties based on decision fusion and near-infrared spectroscopy according to claim 1, characterized in that, Both the first and second quantitative detection models are modeled using partial least squares (PLS). The modeling parameters for the first quantitative detection model are as follows: the sample spectrum is first smoothed using the first derivative of the Savitsky-Golay method with 17 points, and then the near-infrared spectrum in the ranges of 928-1008nm, 1174-1328nm, and 1550-1617nm is extracted for modeling, and the PLS factor number during modeling is 9. The modeling parameters of the second quantitative detection model are as follows: First, the sample spectrum is smoothed using the first derivative of the Savitsky-Golay method with 17 points. Then, the near-infrared spectra at 938nm, 960nm, 1034-1037nm, 1043-1052nm, 1062-1065nm, 1109-1118nm, 1124-1128nm, 1156-1161nm, 1214-1223nm, 1244nm, 1250nm, 1259nm, 1265nm, 1277nm, 1315-1325nm, 1378-1381nm, 1430nm, 1445nm, 1456nm, 1502nm, 1508nm, 1513nm, 1653nm, and 1686nm are used for modeling, and the PLS factor number during modeling is 10.

8. The method for identifying sorghum varieties based on decision fusion and near-infrared spectroscopy according to claim 1, characterized in that, The qualitative discrimination model is modeled using partial least squares method, and the modeling parameters of the qualitative discrimination model are: Near-infrared spectra at 928nm, 996nm, 1087nm, 1223nm, 1322nm, 1345nm, 1669nm, and 1680nm were used for modeling, and the PLS factor was 8 during modeling.

9. The method for identifying sorghum varieties based on decision fusion and near-infrared spectroscopy according to claim 1, characterized in that, The decision fusion model is modeled using discriminant partial least squares (PLS) and the modeling parameter is 3 PLS factors.

10. A sorghum variety discrimination system based on decision fusion and near-infrared spectroscopy, characterized in that, include: The sample collection module is used for the collection and processing of sorghum grain samples. The spectral acquisition module is used to acquire near-infrared spectra for each sample and construct a calibration set; The chemical reference value acquisition module is used to determine the tannin and amylopectin content of each sample; The quantitative model building module is used to build a first quantitative detection model between the near-infrared spectrum of a sample and the tannin content, and to build a second quantitative detection model between the near-infrared spectrum of a sample and the amylopectin content. The qualitative discrimination model building module is used to build a qualitative discrimination model between the near-infrared spectrum of a sample and its class classification. The decision fusion model construction module is used to predict the calibration set using the first quantitative detection model, the second quantitative detection model, and the qualitative discrimination model respectively, to obtain three sets of predicted values. These three sets of predicted values ​​are used as independent variables, and the class label of the sample is used as the dependent variable to construct the decision fusion model. The result prediction module is used to acquire spectra of unknown sorghum samples in the same way as the spectral acquisition module. The first and second quantitative detection models in the quantitative model construction module are used to predict the tannin content and amylopectin content, respectively. The qualitative discriminant model in the qualitative discriminant model construction module is used to predict the category. Then, the decision fusion model in the decision fusion model construction module is used to further predict the above three sets of predicted values ​​to obtain the final category prediction value. When the final category prediction value is ≤0.5, the sample belongs to other varieties, and when the final category prediction value is >0.5, the sample belongs to the target variety.

Images