Method and device for predicting quality of rice product raw material based on partial least squares

Abstract

The application provides a rice product raw material quality prediction method and device based on a partial least squares method, and relates to the technical field of food quality detection and control. The method first divides the rice product raw material into several groups according to weight, extracts equal-weight raw materials from each group as a to-be-detected sample, and uses a Japanese Sakata grain evaluator to detect cracks, yellowing and insect-infested grains in the sample and weigh it. At the same time, the sample is ground into three equal parts, and the Kjeldahl method, oven drying method and iodine colorimetric method are used to respectively determine the content of protein, moisture and amylose in the sample. The infrared spectrum imaging technology is used to record the absorbance of the sample at a specific waveband, and a prediction model of the content of protein, moisture and amylose is established based on the partial least squares method. The crack, yellowing and insect-infested grain proportion are used to construct a rice product raw material quality evaluation system, and the rice product raw material quality is divided into different grades such as excellent, good, qualified and poor according to different evaluation standards.

Description

Technical Field

[0001] This invention relates to the field of food quality testing and control technology, specifically to a method and apparatus for predicting the quality of rice product raw materials based on partial least squares method. Background Technology

[0002] Rice products, a widely consumed food category globally, encompass various forms such as cooked rice, rice noodles, rice cakes, and sushi, and are loved by consumers in different regions. With rising living standards and increasing attention to food safety and quality, higher demands are being placed on the quality control of raw materials for rice products. In the production process, the quality of raw materials directly determines key attributes such as the product's taste, nutritional value, appearance, and shelf life. For example, high-quality rice produces soft, fragrant rice; while rice used for sushi, in addition to taste requirements, has strict requirements for grain integrity and color to ensure the sushi's appearance and quality. For rice noodles, appropriate amylose and protein content affect the noodles' elasticity and texture, preventing problems such as breaking or becoming mushy during cooking.

[0003] Currently, while some technologies and methods exist for testing rice product raw materials, numerous problems remain. Traditional testing methods often rely on subjective sensory judgment, such as visually observing cracked or yellowed grains. This approach is highly subjective, with differing judgment standards among testers leading to poor accuracy and repeatability. Regarding component content testing, some conventional methods are cumbersome and time-consuming. For example, traditional protein content determination methods may require complex chemical digestion and titration processes, failing to meet the need for rapid testing. Moreover, most existing testing technologies target single indicators, lacking a comprehensive analysis and evaluation system for multiple indicators such as protein, moisture, amylose content, and defective particles, making it difficult to comprehensively and accurately reflect the overall quality of rice product raw materials. Faced with complex and diverse rice product raw materials, existing testing and evaluation methods cannot quickly and accurately provide manufacturers with comprehensive raw material quality information. This results in a lack of scientific basis for raw material procurement and production process adjustments, affecting the quality stability and production efficiency of rice products.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method for predicting the quality of rice product raw materials based on partial least squares method, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The method for predicting the quality of rice product raw materials based on partial least squares includes the following steps:

[0008] Step 1: Select samples from rice product raw materials using stratified sampling, collect spectral data using infrared spectroscopy imaging technology, and record the characteristic wavelengths and absorbance related to protein, moisture and amylose.

[0009] Step 2: Determine the protein content of the sample using the Kjeldahl method, the moisture content using the oven drying method, and the amylose content using the iodine colorimetric method. Construct a table of relevant contents based on the characteristic wavelengths and absorbances associated with protein, moisture, and amylose.

[0010] Step 3: Extract the characteristic wavelengths and absorbance of the samples from the relevant content tables, and establish a prediction model for protein, moisture and amylose content based on partial least squares method;

[0011] Step 4: Use stratified sampling estimation method to count yellowed and insect-damaged grains in rice product raw materials, and use prediction model to calculate protein, moisture and amylose content, and construct rice product raw material quality assessment formula for quality prediction.

[0012] Furthermore, the rice product raw materials were divided into 20 equal parts by weight. For each part of rice product raw materials, 10 samples were selected using a simple random sampling method, and the samples were divided into 20 groups.

[0013] Furthermore, infrared spectral imaging technology was used to focus on 20 groups of rice product raw material samples. to Bands, recording bands absorbance ,in, and And set the characteristic wavelength of the protein as The characteristic wavelength of water is and The characteristic wavelength of amylose is .

[0014] Furthermore, after grinding, each group of rice product raw material samples was divided into equal portions by weight. of The protein content of the samples was determined by the Kjeldahl method, the moisture content was determined by the oven drying method, and the amylose content was determined by the iodine colorimetric method.

[0015] The Kjeldahl method for determining the protein content of a sample is as follows:

[0016] The mass to be placed is A sample of rice product raw materials (in grams) was placed in a dry Kjeldahl flask, and then added... Potassium sulfate, Mixtures of copper sulfate and Continue heating the concentrated sulfuric acid until the liquid turns a clear, blue-green color, then continue heating. After the Kjeldahl flask has cooled for several hours, connect it to the Kjeldahl nitrogen distillation apparatus and add it to another conical flask. A mixture of boric acid solution and bromocresol green-methyl red indicator was used. The conical flask was placed under the condenser of the distillation apparatus, with the lower end of the condenser submerged below the surface of the boric acid solution. The solution was then added to the Kjeldahl flask. The sodium hydroxide solution was diluted until the volume of the distillate, i.e., the boric acid solution, reached [a certain value]. Afterwards, use Titration of boric acid solution with hydrochloric acid standard solution resulted in the solution changing from green to grayish-red, at which point the titration was stopped.

[0017] Take another Kjeldahl flask and add... Potassium sulfate, Mixtures of copper sulfate and Continue heating the concentrated sulfuric acid until the liquid turns a clear, blue-green color, then continue heating. After the Kjeldahl flask has cooled for several hours, connect it to the Kjeldahl nitrogen distillation apparatus and add it to another conical flask. A mixture of boric acid solution and bromocresol green-methyl red indicator was used. The conical flask was placed under the condenser of the distillation apparatus, with the lower end of the condenser submerged below the surface of the boric acid solution. The solution was then added to the Kjeldahl flask. The sodium hydroxide solution was diluted until the volume of the distillate, i.e., the boric acid solution, reached [a certain value]. ,use Titrate the boric acid solution with the standard hydrochloric acid solution until the solution turns grayish-red, then stop. Record the volume of standard hydrochloric acid solution consumed, which is the volume of standard hydrochloric acid solution consumed in the blank test.

[0018] Calculation of nitrogen content:

[0019]

[0020] In the formula, This is expressed as the nitrogen content in the sample. This is expressed as the concentration of the hydrochloric acid standard solution. This represents the volume of hydrochloric acid standard solution consumed during sample titration. This represents the volume of standard hydrochloric acid solution consumed in the blank experiment. This is expressed as the quality of the sample. Expressed as the molar mass of nitrogen;

[0021] Protein content conversion:

[0022]

[0023] In the formula, Expressed as protein content;

[0024] The method for determining moisture content using the oven drying method is as follows:

[0025] Select quality as The weighing bottle will have a mass of Place the rice product raw material sample into a weighing bottle and record the total mass. Place the weighing bottle containing the rice product raw material sample into the oven and set the temperature to [temperature value missing]. The drying time is After one hour, remove the weighing bottle and record the mass of the rice product raw material sample and the weighing bottle. Calculate the moisture content:

[0026]

[0027] In the formula, Expressed as moisture content;

[0028] The method for determining amylose content using the iodine colorimetric method is as follows:

[0029] Weigh amylose standard, added After adding anhydrous ethanol, add sodium hydroxide solution Heating in a boiling water bath After cooling for minutes, transfer to In the volumetric flask, use Dilute the water to the mark, then pipette water to the mark separately. Starting with The increments are sequentially increased to... A set of volume data for the solution In a volumetric flask, dilute to volume with distilled water to a concentration of [value missing]. Starting with The increments are sequentially increased to... A series of standard solutions with a set of concentration data were prepared. Amylopectin standard was weighed and prepared according to the above method for preparing amylose standard solutions to a concentration of [missing value]. The solution was then diluted with the same gradient to prepare a series of standard solutions of different concentrations; weigh out... Dissolve potassium iodide in a small amount of distilled water, add After the iodine has completely dissolved, dilute to the final volume with distilled water. Store in a brown bottle; weigh of Sodium hydroxide solution, dissolved in distilled water, cooled and then diluted to a final volume. ;

[0030] quality is Rice product raw material samples were placed Add to the conical flask Stir well with anhydrous ethanol, soak for 1 hour, filter, cool to room temperature, and transfer the solution to... In a volumetric flask, dilute to the mark with distilled water and shake well to obtain a sample solution of rice product raw materials. Then, pipette different concentrations of amylose standard solutions. At Add to each colorimetric tube Iodine reagent, diluted to volume with distilled water, shaken well and then left to stand. Minutes, with distilled water as a blank control, at wavelength... At the same time, the absorbance of each standard solution was measured by spectrophotometer. A standard curve was plotted with the amylose concentration as the abscissa and the absorbance as the ordinate, and the regression equation of amylose was obtained.

[0031] draw Rice product raw material sample solution in In the colorimetric tube, iodine reagent was added and diluted to volume according to the steps for plotting the standard curve. The solution was then shaken well and allowed to stand for 15 minutes to obtain the raw material sample of the tested rice product. The absorbance of the sample solution was measured at a specific wavelength; subsequently, a blank test was performed without adding any sample, following the sample measurement procedure, and the absorbance was measured. Blank absorbance at the specified wavelength;

[0032] exist At a given wavelength, the difference between the absorbance of the rice product raw material sample solution and the blank absorbance was calculated. This difference was then compared with the amylose concentration obtained from the amylose regression equation, and the dilution factor was calculated.

[0033]

[0034] In the formula, Expressed as a dilution factor;

[0035] The formula for calculating amylose content is:

[0036]

[0037] In the formula, Expressed as amylose content, Expressed as amylose concentration obtained from the amylose regression equation, in units of... , The volume of the solution in the test rice product raw material sample is expressed as [volume unit]. .

[0038] Furthermore, a univariate linear regression equation is constructed:

[0039]

[0040] In the formula, Expressed as absorbance. Expressed as amylose concentration, Represented as intercept, Expressed as slope;

[0041] The absorbance of standard solutions of amylose at different concentrations was measured experimentally to obtain a series of corresponding amylose concentrations. and absorbance data points , recorded as ;

[0042] Calculate the average concentration of amylose:

[0043]

[0044] In the formula, This is expressed as the average concentration of amylose.

[0045] Calculate the average absorbance of standard solutions of amylose at different concentrations:

[0046]

[0047] In the formula, The absorbance is expressed as the average value of the standard amylose solutions of different concentrations.

[0048] Calculate the slope of the univariate linear regression equation:

[0049]

[0050] Calculate the intercept of the univariate linear regression equation:

[0051] .

[0052] Furthermore, based on the extracted 20 sets of data, each set containing the absorbance of 5 characteristic wavelengths, the dimension was set as follows: Absorbance data matrix:

[0053]

[0054] In the formula, Represented as an absorbance data matrix;

[0055] Set the dimension as Dependent variable matrix:

[0056]

[0057] In the formula, Represented as a dependent variable matrix;

[0058] Calculate the dependent variable matrix The average of columns 1, 2, and 3:

[0059]

[0060]

[0061]

[0062] In the formula, and Represented as dependent variable matrices The average of columns 1, 2, and 3. Represented as the row index of the matrix;

[0063] Calculate the dependent variable matrix Standard deviations in columns 1, 2, and 3:

[0064]

[0065]

[0066]

[0067] In the formula, and Represented as a dependent variable matrix Standard deviations in columns 1, 2, and 3;

[0068] For absorbance data matrix and dependent variable matrix Standardization process:

[0069]

[0070] In the formula, Represented as raw data, where, Represented as column indices of a matrix. , Represented as an absorbance data matrix column index, Represented as a dependent variable matrix column index, Represented as the first The average of the column, Represented as the first The standard deviation of the column is used to substitute the standardized result back into the original absorbance data matrix. and dependent variable matrix ;

[0071] Among them, the The average value of the column is:

[0072]

[0073] No. The standard deviation of the column is:

[0074]

[0075] Calculate the absorbance data matrix and dependent variable matrix Covariance matrix:

[0076]

[0077] In the formula, Represented as an absorbance data matrix and dependent variable matrix The covariance matrix;

[0078] Set the dimension as The process matrix is ​​as follows:

[0079]

[0080] In the formula, Represented as a process matrix;

[0081] Set the dimension as The process matrix is ​​as follows:

[0082]

[0083] In the formula, Represented as a process two matrix;

[0084] Calculate eigenvalues ​​and eigenvectors:

[0085] because ,have to:

[0086]

[0087] In the formula, Represented as eigenvalues, Represented as an eigenvector, Represented as dimension The identity matrix;

[0088] The characteristic equation of the matrix in the construction process:

[0089]

[0090] In the formula, The characteristic equation of the process two matrix is ​​expressed as follows: eigenvalues ;

[0091] For each eigenvalue Solve the homogeneous linear equation system The corresponding feature vectors are obtained. ;

[0092] contrast eigenvalues Given the size of the eigenvalues, find the largest eigenvalue and denote it as . Its corresponding eigenvector is denoted as ;

[0093] Next, calculate the number of principal components to be extracted:

[0094] Set the number of principal components to be For the first The variance contribution rate of each principal component is:

[0095]

[0096] In the formula, Represented as the first The variance contribution rate of each principal component;

[0097] forward The formula for calculating the cumulative variance contribution rate of each principal component is:

[0098]

[0099] Starting with the first principal component, calculate the cumulative variance contribution rate sequentially. When it first reaches or exceeds 0.85, That is, the number of principal components that are determined;

[0100] Calculate the absorbance data matrix First component:

[0101]

[0102] In the formula, Represented as an absorbance data matrix The first component;

[0103] Calculate the dependent variable right Regression coefficients:

[0104]

[0105] In the formula, Represented as dependent variable right The regression coefficients;

[0106] Calculate the dependent variable matrix First component:

[0107]

[0108] In the formula, Represented as a dependent variable matrix The first component;

[0109] Calculate the absorbance data matrix The residual matrix:

[0110]

[0111] In the formula, Represented as an absorbance data matrix The residual matrix, Represented as the first load vector, the calculation formula is:

[0112]

[0113] Calculate the dependent variable matrix The residual matrix:

[0114]

[0115] In the formula, Represented as a dependent variable matrix The residual matrix;

[0116] For absorbance data matrix residual matrix and dependent variable matrix residual matrix Repeat the above steps until the first one is extracted. For ingredients and :

[0117] calculate Covariance matrix:

[0118]

[0119] In the formula, Represented as covariance matrix

[0120] Calculate the first The residual matrix at the next iteration and Cross-covariance feature matrix:

[0121]

[0122] In the formula, Represented as the cross-covariance feature matrix;

[0123] Perform eigenvalue decomposition on it, and select the eigenvector corresponding to the largest eigenvalue, denoted as . ,calculate The One component:

[0124]

[0125] calculate right Regression coefficients:

[0126]

[0127] In the formula, Represented as right The regression coefficients;

[0128] calculate The One component:

[0129]

[0130] Calculate the first Residual matrix of the extracted components, Part 1:

[0131]

[0132] In the formula, Represented as the first The residual matrix of the extracted components, Represented as the first There are load vectors, where...

[0133]

[0134] Calculate the first Residual matrix of the second extracted component:

[0135]

[0136] In the formula, Represented as the first The second residual matrix of the extracted components;

[0137] Constructing the component matrix:

[0138]

[0139] In the formula, Represented as a component matrix;

[0140] Constructing the regression coefficient matrix:

[0141]

[0142] In the formula, Represented as a regression coefficient matrix;

[0143] Construct a standardized regression model:

[0144]

[0145] because ,set up

[0146] The standardized prediction model is as follows:

[0147]

[0148] Standardized prediction models Denormalized, the protein prediction model is as follows:

[0149]

[0150] In the formula, Represented as a predictive model for proteins;

[0151] The moisture prediction model is as follows:

[0152]

[0153] In the formula, Represented as a moisture prediction model;

[0154]

[0155] In the formula, This is represented as a prediction model for amylose.

[0156] Furthermore, the rice product raw materials are divided into equal weight portions. Groups, draw equal weight from each group Rice raw materials were used as the test samples. A Satake grain size analyzer (Japan) was used to identify cracked, yellowed, and insect-damaged grains in each sample. The identified cracked, yellowed, and insect-damaged grains were weighed separately, and their weights were recorded. ,in, ;

[0157] Calculate the percentage of cracked, yellowed, and insect-damaged particles in each sample group:

[0158]

[0159] In the formula, Represented as the first The proportion of cracked particles, yellowed particles, and insect-damaged particles;

[0160] Calculate the percentage of cracked grains, yellowed grains, and insect-damaged grains in the raw materials for rice products:

[0161]

[0162] In the formula, This represents the percentage of cracked grains, yellowed grains, and insect-damaged grains in the raw materials for rice products.

[0163] Furthermore, for Group of samples to be tested, predict the first ( Represented as protein, Represented as moisture. The method for expressing the content of amylose (amylose) is as follows:

[0164] Extract the bands respectively The absorbance, the construction dimension is The absorbance data matrix to be detected :

[0165] The absorbance data matrix to be detected Standardization process:

[0166]

[0167] In the formula, Represented as a row index, Represented as a column index, Represented as the first element of the normalized absorbance data matrix to be detected. OK, The data in the column were used to obtain a standardized absorbance data matrix for testing. ;

[0168] The computational dimension is The new sample was extracted Projection matrices along the directions of the principal components:

[0169]

[0170] In the formula, This is represented as a new sample in the extraction. Projection matrices along the directions of each principal component;

[0171] The projection of the new sample onto the principal component direction after standardization is:

[0172]

[0173] This is represented as the projection of the new sample onto the principal component direction after standardization. This represents the index of the principal component, where, , Represented as The Middle Line number Elements of column index, Represented as The Middle Line number Elements of the column index;

[0174]

[0175] In the formula, Represented as The matrix, Represented as dimension Estimates of the regression coefficient matrix;

[0176] Calculate the first The new sample in the first Standardized intermediate predicted values ​​of seed content:

[0177]

[0178] In the formula, Represented as the first The first sample to be tested Standardized intermediate predicted values ​​of species content, Represented as The Middle Line number Column elements;

[0179] Calculate the first The nth sample to be tested was obtained after principal component projection and regression coefficient calculation. The standardized prediction formula for the content of a species is:

[0180]

[0181] In the formula, Represented as the first The first sample to be tested Standardized predicted values ​​of the content of each component. Represented as the first The sample to be tested was in the first... The residual in the prediction of the content of each component is set to 0.

[0182] Substituting into the prediction model, then the first... The first new sample The formula for predicting the content of each component is:

[0183]

[0184] In the formula, Represented as the first The first new sample Predicted values ​​of component content Represented as a dependent variable matrix The Middle The standard deviation of the column, Represented as a dependent variable matrix The Middle The average of the column;

[0185] The protein content of rice product raw materials is:

[0186]

[0187] In the formula, This refers to the protein content of rice-based raw materials;

[0188] The moisture content of rice product raw materials is:

[0189]

[0190] In the formula, This refers to the moisture content of the raw materials used in rice products.

[0191] The amylose content of rice product raw materials is:

[0192]

[0193] In the formula, This refers to the amylose content of rice product raw materials.

[0194] Furthermore, based on the protein, moisture, and amylose content of rice product raw materials, and the proportions of cracked, yellowed, and insect-damaged grains, a quality assessment formula for rice product raw materials is constructed:

[0195]

[0196] In the formula, This is expressed as a formula for evaluating the quality of rice product raw materials. and Represented as weighting coefficients, and , This is represented as an adjustment parameter;

[0197] Set the quality threshold as and ,and ,when At that time, the quality assessment of rice product raw materials was excellent. At that time, the quality assessment of rice product raw materials was good. At that time, the quality assessment of rice product raw materials was qualified. At that time, the quality assessment of rice product raw materials was poor.

[0198] Compared with the prior art, the beneficial effects of the present invention are:

[0199] This invention incorporates key indicators of rice product raw materials, such as protein, moisture, amylose content, and the presence of cracked, yellowed, and insect-damaged grains, into a unified evaluation system. Through scientific testing procedures and professional equipment, it comprehensively and accurately grasps the various characteristics of the raw materials, avoiding the limitations of traditional single-indicator testing. The content prediction model, based on partial least squares, possesses powerful data processing capabilities, enabling in-depth analysis of the complex nonlinear relationships between protein, moisture, amylose content, and numerous influencing factors in rice product raw materials. This allows for accurate assessment of the impact of each factor on protein, moisture, and amylose content, providing strong support for precise prediction. By strictly controlling raw material quality, it provides a stable and reliable raw material foundation for rice product production, ensuring consistently high quality in terms of taste, nutritional value, appearance, and shelf life.

[0200] The invention also provides an apparatus for predicting the quality of rice product raw materials based on partial least squares method. The apparatus is used to execute the aforementioned method for predicting the quality of rice product raw materials based on partial least squares method, comprising:

[0201] The spectral data acquisition module is used to select samples from rice product raw materials using a stratified sampling method, acquire spectral data using infrared spectral imaging technology, and record characteristic wavelengths and absorbances related to protein, moisture and amylose.

[0202] The content determination module is used to determine the protein content of a sample using the Kjeldahl nitrogen determination method, the moisture content using the oven drying method, and the amylose content using the iodine colorimetric method. It also constructs a content table based on the characteristic wavelengths and absorbances associated with protein, moisture, and amylose.

[0203] The model building module is used to extract the characteristic wavelengths and absorbance of samples from the relevant content tables, and to build predictive models for protein, moisture and amylose content based on partial least squares method.

[0204] The quality prediction module is used to statistically analyze yellowed and insect-damaged grains in rice product raw materials using stratified sampling estimation method, and to calculate the protein, moisture and amylose content using prediction model, and to construct a quality assessment formula for rice product raw materials for quality prediction. Attached Figure Description

[0205] Figure 1 This is a schematic diagram of the overall method flow of the present invention;

[0206] Figure 2 This is a schematic diagram of the overall system modules of the present invention. Detailed Implementation

[0207] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0208] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0209] Example:

[0210] Please see Figure 1 The present invention provides a technical solution:

[0211] The method for predicting the quality of rice product raw materials based on partial least squares includes the following steps:

[0212] Step 1: Select samples from rice product raw materials using stratified sampling, collect spectral data using infrared spectroscopy imaging technology, and record the characteristic wavelengths and absorbance related to protein, moisture and amylose.

[0213] Using precise weighing equipment, the rice product raw materials are divided into 20 equal portions by weight, each portion having the same mass to ensure consistency and fairness in sampling. For example, if the total mass of the rice product raw materials is 100 kg, then each portion weighs 5 kg. For each portion of rice product raw materials, a simple random sampling method is used to select 10 samples, and these 10 samples are grouped into one group, resulting in a total of 20 groups of samples.

[0214] 1500 cm -1Nearby, protein molecules contain numerous amide bonds. In infrared spectroscopy, the coupling of the NH bending vibration and CN stretching vibration of these amide bonds produces a characteristic absorption peak, known as the amide II band, which can be used to characterize the presence and abundance of proteins. (1700 cm⁻¹) -1 Near the amide I band, the stretching vibration of the carbonyl group (C=O) in the protein will produce an absorption peak. Due to the interaction of different amino acid residues in the protein molecule and the secondary structure of the protein, the position of the amide I band will change. It is one of the important characteristic wavelengths of the protein.

[0215] 1600cm -1 Nearby, the HOH bending vibration in water molecules produces an absorption peak. This is because the water molecule undergoes a change in its molecular dipole moment under this vibrational mode, thus absorbing infrared light of a corresponding wavelength, which is a characteristic absorption wavelength of water. 3200 cm⁻¹ -1 Nearby: The stretching vibration of the OH group in water molecules produces a strong absorption peak. Due to the hydrogen bonding between water molecules, the absorption peak of the OH stretching vibration is broadened and its position is shifted to 3200 cm⁻¹. -1 There is obvious absorption at this point, which can be used as an important characteristic wavelength for detecting moisture.

[0216] 1000 cm -1 Nearby, amylose is a linear polysaccharide composed of glucose units linked by α-1,4-glycosidic bonds. In infrared spectroscopy, amylose appears at 1000 cm⁻¹. -1 It has a relatively unique absorption peak, which is related to the structure and bond vibration characteristics of glucose units in amylose molecules. This absorption peak can be used for qualitative and quantitative analysis of amylose.

[0217] Therefore, infrared spectral imaging technology was used to focus on 20 groups of rice product raw material samples. to Band, resolution set to Recording bands absorbance ,in, and And set the characteristic wavelength of the protein as The characteristic wavelength of water is and The characteristic wavelength of amylose is .

[0218] Step 2: Determine the protein content of the sample using the Kjeldahl method, the moisture content using the oven drying method, and the amylose content using the iodine colorimetric method. Construct a table of relevant contents based on the characteristic wavelengths and absorbances associated with protein, moisture, and amylose.

[0219] The Kjeldahl method, based on the relatively stable nitrogen content in proteins, converts organic nitrogen in the sample into inorganic nitrogen for quantitative analysis, thereby calculating the protein content. This method boasts high accuracy and precision, making it a classic and widely accepted method for protein content determination. The oven drying method, based on the principle of moisture evaporation, removes moisture from the sample under specific temperature and time conditions, calculating the moisture content based on the mass difference before and after drying. Amylose forms a specific blue complex with iodine, the intensity of which is directly proportional to the amylose content. The iodine colorimetric method utilizes this property, determining amylose content by measuring absorbance. This method exhibits high specificity for amylose, accurately determining its content in rice products without interference from other polysaccharides.

[0220] After grinding, each group of rice product raw material samples was divided into equal portions with a mass of [missing information]. of This is because rice product raw materials may have differences in particle size, component distribution, etc., and even after grinding, it is difficult to guarantee complete uniformity. Dividing each sample equally ensures that the samples used for different index determinations come from the "same" sample with the same or similar components, structure, and characteristics, reducing the fluctuation of results caused by differences in the samples themselves. For example, if they are not divided equally, the sample used to determine protein may contain more particles with high protein content, while the sample used to determine amylose may contain fewer such particles, which would make the determination results not truly reflect the overall situation of the rice product raw materials. The protein content of the samples was determined by the Kjeldahl method, the moisture content by the oven drying method, and the amylose by the iodine colorimetric method.

[0221] The Kjeldahl method for determining the protein content of a sample is as follows:

[0222] The mass to be placed is A sample of rice product raw materials (in grams) was placed in a dry Kjeldahl flask, and then added... Potassium sulfate, Mixtures of copper sulfate and Continue heating the concentrated sulfuric acid until the liquid turns a clear, blue-green color, then continue heating. After the Kjeldahl flask has cooled for several hours, connect it to the Kjeldahl nitrogen distillation apparatus and add it to another conical flask. A mixture of boric acid solution and bromocresol green-methyl red indicator was used. The conical flask was placed under the condenser of the distillation apparatus, with the lower end of the condenser submerged below the surface of the boric acid solution. The solution was then added to the Kjeldahl flask. The sodium hydroxide solution was diluted until the volume of the distillate, i.e., the boric acid solution, reached [a certain value]. Afterwards, use Titration of boric acid solution with hydrochloric acid standard solution resulted in the solution changing from green to grayish-red, at which point the titration was stopped.

[0223] Take another Kjeldahl flask and add... Potassium sulfate, Mixtures of copper sulfate and Continue heating the concentrated sulfuric acid until the liquid turns a clear, blue-green color, then continue heating. After the Kjeldahl flask has cooled for several hours, connect it to the Kjeldahl nitrogen distillation apparatus and add it to another conical flask. A mixture of boric acid solution and bromocresol green-methyl red indicator was used. The conical flask was placed under the condenser of the distillation apparatus, with the lower end of the condenser submerged below the surface of the boric acid solution. The solution was then added to the Kjeldahl flask. The sodium hydroxide solution was diluted until the volume of the distillate, i.e., the boric acid solution, reached [a certain value]. ,use Titrate the boric acid solution with the standard hydrochloric acid solution until the solution turns grayish-red, then stop. Record the volume of standard hydrochloric acid solution consumed, which is the volume of standard hydrochloric acid solution consumed in the blank test.

[0224] Calculation of nitrogen content:

[0225]

[0226] In the formula, This is expressed as the nitrogen content in the sample. This is expressed as the concentration of the hydrochloric acid standard solution. This represents the volume of hydrochloric acid standard solution consumed during sample titration. This represents the volume of standard hydrochloric acid solution consumed in the blank experiment. This is expressed as the quality of the sample. Expressed as the molar mass of nitrogen;

[0227] Protein content conversion:

[0228]

[0229] In the formula, This refers to protein content. Proteins are biological macromolecules composed of amino acids linked by peptide bonds. Generally, the nitrogen content in proteins is relatively stable, averaging approximately [missing information]. This is a statistical average obtained through the analysis and measurement of a large number of proteins from different sources and of different types. ;

[0230] The method for determining moisture content using the oven drying method is as follows:

[0231] Select quality as The weighing bottle will have a mass of Place the rice product raw material sample into a weighing bottle and record the total mass. Place the weighing bottle containing the rice product raw material sample into the oven and set the temperature to [temperature value missing]. The drying time is After one hour, remove the weighing bottle and record the mass of the rice product raw material sample and the weighing bottle. Calculate the moisture content:

[0232]

[0233] In the formula, Expressed as moisture content;

[0234] The method for determining amylose content using the iodine colorimetric method is as follows:

[0235] Weigh amylose standard, added After adding anhydrous ethanol, add sodium hydroxide solution Heating in a boiling water bath After cooling for minutes, transfer to In the volumetric flask, use Dilute the water to the mark, then pipette water to the mark separately. Starting with The increments are sequentially increased to... A set of volume data for the solution In a volumetric flask, dilute to volume with distilled water to a concentration of [value missing]. Starting with The increments are sequentially increased to... A series of standard solutions with a set of concentration data were prepared. Amylopectin standard was weighed and prepared according to the above method for preparing amylose standard solutions to a concentration of [missing value]. The solution was then diluted with the same gradient to prepare a series of standard solutions of different concentrations; weigh out... Dissolve potassium iodide in a small amount of distilled water, add After the iodine has completely dissolved, dilute to the final volume with distilled water. Store in a brown bottle; weigh of Sodium hydroxide solution, dissolved in distilled water, cooled and then diluted to a final volume. ;

[0236] quality is Rice product raw material samples were placed Add to the conical flask Stir well with anhydrous ethanol, soak for 1 hour, filter, cool to room temperature, and transfer the solution to... In a volumetric flask, dilute to the mark with distilled water and shake well to obtain a sample solution of rice product raw materials. Then, pipette different concentrations of amylose standard solutions. At Add to each colorimetric tube Iodine reagent, diluted to volume with distilled water, shaken well and then left to stand. Minutes, with distilled water as a blank control, at wavelength... At the same time, the absorbance of each standard solution was measured by spectrophotometer. A standard curve was plotted with the amylose concentration as the abscissa and the absorbance as the ordinate, and the regression equation of amylose was obtained.

[0237] draw Rice product raw material sample solution in In the colorimetric tube, iodine reagent was added and diluted to volume according to the steps for plotting the standard curve. The solution was then shaken well and allowed to stand for 15 minutes to obtain the raw material sample of the tested rice product. The absorbance of the sample solution was measured at a specific wavelength; subsequently, a blank test was performed without adding any sample, following the sample measurement procedure, and the absorbance was measured. Blank absorbance at the specified wavelength;

[0238] exist At a given wavelength, the difference between the absorbance of the rice product raw material sample solution and the blank absorbance was calculated. This difference was then compared with the amylose concentration obtained from the amylose regression equation, and the dilution factor was calculated.

[0239]

[0240] In the formula, Expressed as a dilution factor;

[0241] The formula for calculating amylose content is:

[0242]

[0243] In the formula, Expressed as amylose content, Expressed as amylose concentration obtained from the amylose regression equation, in units of... , The volume of the solution in the test rice product raw material sample is expressed as [volume unit]. .

[0244] The steps for constructing the regression equation for amylose are as follows:

[0245] Construct a univariate linear regression equation:

[0246]

[0247] In the formula, Expressed as absorbance. Expressed as amylose concentration, Represented as intercept, Expressed as slope;

[0248] The absorbance of standard solutions of amylose at different concentrations was measured experimentally to obtain a series of corresponding amylose concentrations. and absorbance data points , recorded as ;

[0249] Calculate the average concentration of amylose:

[0250]

[0251] In the formula, This is expressed as the average concentration of amylose.

[0252] Calculate the average absorbance of standard solutions of amylose at different concentrations:

[0253]

[0254] In the formula, The absorbance is expressed as the average value of the standard amylose solutions of different concentrations.

[0255] Calculate the slope of the univariate linear regression equation:

[0256]

[0257] Calculate the intercept of the univariate linear regression equation:

[0258]

[0259] Construct a table of relevant contents, where the row headings include sample number, characteristic wavelength and corresponding absorbance of protein, characteristic wavelength and corresponding absorbance of water, and characteristic wavelength of amylose. According to the grouping order of the samples, number each group of samples from 1 to 20 and fill it in the first column of the table. For each group of samples, fill in the characteristic wavelengths of protein, water and amylose and their respective absorbances in the corresponding row.

[0260] Step 3: Extract the characteristic wavelengths and absorbance of the samples from the relevant content tables, and establish a prediction model for protein, moisture and amylose content based on partial least squares method;

[0261] When studying protein, moisture, and amylose content, multiple influencing factors and variables are often involved. Partial least squares (PLS) can effectively handle the complex relationships between multiple independent and dependent variables, comprehensively considering the influence of various factors on the content, unlike some traditional methods that may only be able to handle a single variable or a simple linear relationship.

[0262] In actual measurements, the independent variables used to predict protein, moisture, and amylose content are highly correlated, i.e., multicollinearity. Partial least squares method can eliminate or reduce the effects of multicollinearity among variables by extracting principal components, making the model more stable and accurate, and able to accurately assess the true contribution of each variable to the content.

[0263] The characteristic wavelengths and absorbance of the samples were extracted from the relevant content tables. Based on the extracted 20 sets of data, each set contained the absorbance of 5 characteristic wavelengths, and the dimension was set as follows. Absorbance data matrix:

[0264]

[0265] In the formula, Represented as an absorbance data matrix;

[0266] Set the dimension as Dependent variable matrix:

[0267]

[0268] In the formula, Represented as a dependent variable matrix;

[0269] Calculate the dependent variable matrix The average of columns 1, 2, and 3:

[0270]

[0271]

[0272]

[0273] In the formula, and Represented as dependent variable matrices The average of columns 1, 2, and 3. Represented as the row index of the matrix;

[0274] Calculate the dependent variable matrix Standard deviations in columns 1, 2, and 3:

[0275]

[0276]

[0277]

[0278] In the formula, and Represented as a dependent variable matrix Standard deviations in columns 1, 2, and 3;

[0279] For absorbance data matrix and dependent variable matrix Standardization process:

[0280]

[0281] In the formula, Represented as raw data, where, Represented as column indices of a matrix. , Represented as an absorbance data matrix column index, Represented as a dependent variable matrix column index, Represented as the first The average of the column, Represented as the first The standard deviation of the column is used to substitute the standardized result back into the original absorbance data matrix. and dependent variable matrix ;

[0282] Among them, the The average value of the column is:

[0283]

[0284] No. The standard deviation of the column is:

[0285]

[0286] Calculate the absorbance data matrix and dependent variable matrix Covariance matrix:

[0287]

[0288] In the formula, Represented as an absorbance data matrix and dependent variable matrix The covariance matrix;

[0289] Set the dimension as The process matrix is ​​as follows:

[0290]

[0291] In the formula, Represented as a process matrix;

[0292] Set the dimension as The process matrix is ​​as follows:

[0293]

[0294] In the formula, Represented as a process two matrix;

[0295] Calculate eigenvalues ​​and eigenvectors:

[0296] because ,have to:

[0297]

[0298] In the formula, Represented as eigenvalues, Represented as an eigenvector, Represented as dimension The identity matrix;

[0299] The characteristic equation of the matrix in the construction process:

[0300]

[0301] In the formula, The characteristic equation of the process two matrix is ​​expressed as follows: eigenvalues ;

[0302] For each eigenvalue Solve the homogeneous linear equation system The corresponding feature vectors are obtained. ;

[0303] contrast eigenvalues Given the size of the eigenvalues, find the largest eigenvalue and denote it as . Its corresponding eigenvector is denoted as ;

[0304] Next, calculate the number of principal components to be extracted:

[0305] Set the number of principal components to be For the first The variance contribution rate of each principal component is:

[0306]

[0307] In the formula, Represented as the first The variance contribution rate of each principal component;

[0308] forward The formula for calculating the cumulative variance contribution rate of each principal component is:

[0309]

[0310] Starting with the first principal component, calculate the cumulative variance contribution rate sequentially. When it first reaches or exceeds 0.85, That is, the number of principal components that are determined;

[0311] Calculate the absorbance data matrix First component:

[0312]

[0313] In the formula, Represented as an absorbance data matrix The first component;

[0314] Calculate the dependent variable right Regression coefficients:

[0315]

[0316] In the formula, Represented as dependent variable right The regression coefficients;

[0317] Calculate the dependent variable matrix First component:

[0318]

[0319] In the formula, Represented as a dependent variable matrix The first component;

[0320] Calculate the absorbance data matrix The residual matrix:

[0321]

[0322] In the formula, Represented as an absorbance data matrix The residual matrix, Represented as the first load vector, the calculation formula is:

[0323]

[0324] Calculate the dependent variable matrix The residual matrix:

[0325]

[0326] In the formula, Represented as a dependent variable matrix The residual matrix;

[0327] For absorbance data matrix residual matrix and dependent variable matrix residual matrix Repeat the above steps until the first one is extracted. For ingredients and :

[0328] calculate Covariance matrix:

[0329]

[0330] In the formula, Represented as covariance matrix

[0331] Calculate the first The residual matrix at the next iteration and Cross-covariance feature matrix:

[0332]

[0333] In the formula, Represented as the cross-covariance feature matrix;

[0334] Perform eigenvalue decomposition on it, and select the eigenvector corresponding to the largest eigenvalue, denoted as . ,calculate The One component:

[0335]

[0336] calculate right Regression coefficients:

[0337]

[0338] In the formula, Represented as right The regression coefficients;

[0339] calculate The One component:

[0340]

[0341] Calculate the first Residual matrix of the extracted components, Part 1:

[0342]

[0343] In the formula, Represented as the first The residual matrix of the extracted components, Represented as the first There are load vectors, where...

[0344]

[0345] Calculate the first Residual matrix of the second extracted component:

[0346]

[0347] In the formula, Represented as the first The second residual matrix of the extracted components;

[0348] Constructing the component matrix:

[0349]

[0350] In the formula, Represented as a component matrix;

[0351] Constructing the regression coefficient matrix:

[0352]

[0353] In the formula, Represented as a regression coefficient matrix;

[0354] Construct a standardized regression model:

[0355]

[0356] because ,set up

[0357] The standardized prediction model is as follows:

[0358]

[0359] Standardized prediction models Denormalized, the protein prediction model is as follows:

[0360]

[0361] In the formula, Represented as a predictive model for proteins;

[0362] The moisture prediction model is as follows:

[0363]

[0364] In the formula, Represented as a moisture prediction model;

[0365]

[0366] In the formula, This is represented as a prediction model for amylose.

[0367] Step 4: Use stratified sampling estimation method to count yellowed and insect-damaged grains in rice product raw materials, and use prediction model to calculate protein, moisture and amylose content, and construct rice product raw material quality assessment formula for quality prediction.

[0368] Rice product raw materials are divided into equal weight categories After each group, draw an equal weight from each group. Rice raw materials were used as the samples to be tested. This ensures that the sampled material evenly covers the entire rice product raw material group, avoiding the possibility that some raw materials will not be detected due to sampling bias. This makes the test results more accurately reflect the overall quality of the raw materials. The Satake Particle Evaluator from Japan was used to identify cracked, yellowed, and insect-damaged grains in each sample of rice raw material. The identified cracked, yellowed, and insect-damaged grains were weighed separately, and their weights were recorded. ,in, In the food industry, there are strict standards and regulations for the quality of rice raw materials. Conducting such tests can ensure that the raw materials used by enterprises meet the relevant standards and avoid reputational damage due to raw material quality issues.

[0369] Calculate the percentage of cracked, yellowed, and insect-damaged particles in each sample group:

[0370]

[0371] In the formula, Represented as the first The proportion of cracked particles, yellowed particles, and insect-damaged particles;

[0372] Calculate the percentage of cracked grains, yellowed grains, and insect-damaged grains in the raw materials for rice products:

[0373]

[0374] In the formula, This represents the percentage of cracked grains, yellowed grains, and insect-damaged grains in the raw materials for rice products.

[0375] for Group of samples to be tested, predict the first ( Represented as protein, Represented as moisture. The method for expressing the content of amylose (amylose) is as follows:

[0376] Extract the bands respectively The absorbance, the construction dimension is The absorbance data matrix to be detected :

[0377] The absorbance data matrix to be detected Standardization process:

[0378]

[0379] In the formula, Represented as a row index, Represented as a column index, Represented as the first element of the normalized absorbance data matrix to be detected. OK, The data in the column were used to obtain a standardized absorbance data matrix for testing. ;

[0380] The computational dimension is The new sample was extracted Projection matrices along the directions of the principal components:

[0381]

[0382] In the formula, This is represented as a new sample in the extraction. Projection matrices along the directions of each principal component;

[0383] The projection of the new sample onto the principal component direction after standardization is:

[0384]

[0385] This is represented as the projection of the new sample onto the principal component direction after standardization. This represents the index of the principal component, where, , Represented as The Middle Line number Elements of column index, Represented as The Middle Line number Elements of the column index;

[0386]

[0387] In the formula, Represented as The matrix, Represented as dimension Estimates of the regression coefficient matrix;

[0388] Calculate the first The new sample in the first Standardized intermediate predicted values ​​of seed content:

[0389]

[0390] In the formula, Represented as the first The first sample to be tested Standardized intermediate predicted values ​​of species content, Represented as The Middle Line number Column elements;

[0391] Calculate the first The nth sample to be tested was obtained after principal component projection and regression coefficient calculation. The standardized prediction formula for the content of a species is:

[0392]

[0393] In the formula, Represented as the first The first sample to be tested Standardized predicted values ​​of the content of each component. Represented as the first The sample to be tested was in the first... The residual in the prediction of the content of each component is set to 0.

[0394] Substituting into the prediction model, then the first... The first new sample The formula for predicting the content of each component is:

[0395]

[0396] In the formula, Represented as the first The first new sample Predicted values ​​of component content Represented as a dependent variable matrix The Middle The standard deviation of the column, Represented as a dependent variable matrix The Middle The average of the column;

[0397] The protein content of rice product raw materials is:

[0398]

[0399] In the formula, This refers to the protein content of rice-based raw materials;

[0400] The moisture content of rice product raw materials is:

[0401]

[0402] In the formula, This refers to the moisture content of the raw materials used in rice products.

[0403] The amylose content of rice product raw materials is:

[0404]

[0405] In the formula, Indicates the amylose content of rice product raw materials;

[0406] For example, the method for calculating the protein content of a sample is as follows:

[0407] for Groups of samples to be tested, each with extracted wavelengths as follows: The absorbance, the construction dimension is The absorbance data matrix to be detected :

[0408] The absorbance data matrix to be detected Standardization process:

[0409]

[0410] The computational dimension is The new sample was extracted Projection matrices along the directions of the principal components:

[0411]

[0412] The projection of the new sample onto the principal component direction after standardization is:

[0413]

[0414]

[0415] Calculate the first Standardized intermediate predicted values ​​of protein content in a new sample:

[0416]

[0417] In the formula, Represented as the first Standardized intermediate predicted values ​​of protein content in a new sample. Represented as The Middle The element in the first column of the row;

[0418] Calculate the first The formula for the standardized predicted value of protein content obtained by principal component projection and regression coefficient calculation of a sample to be tested is:

[0419]

[0420] In the formula, Represented as the first The standardized predicted protein content of each sample to be tested is obtained by calculating the principal component projection and regression coefficients. Represented as the first The residual in the protein content prediction of each sample to be tested is set to 0.

[0421] Substituting into the protein prediction model, then the first... The formula for predicting the protein content of a new sample is:

[0422] .

[0423] Subsequently, based on the protein, moisture, and amylose content of rice product raw materials, as well as the proportions of cracked, yellowed, and insect-damaged grains, a quality assessment formula for rice product raw materials was constructed:

[0424]

[0425] In the formula, This is expressed as a formula for evaluating the quality of rice product raw materials. and These are represented as weighting coefficients, which need to be set according to different actual situations, and , This is an adjustable parameter, with a default range of 1-10;

[0426] The weighting coefficients reflect the relative importance of various factors to the quality of rice product raw materials. They can be adjusted according to individual needs and product characteristics. For example, in making high-grade sushi rice, a high amylose content results in a drier texture and harder rice, while a low amylose content leads to a softer, stickier texture. Moisture content affects water absorption and expansion during cooking, affecting the overall texture. It is recommended to set the weighting coefficients to [value missing]. .

[0427] Set the quality threshold as and ,and ,set up ,when At that time, the quality assessment of rice product raw materials was excellent. At that time, the quality assessment of rice product raw materials was good. At that time, the quality assessment of rice product raw materials was qualified. At that time, the quality assessment of rice product raw materials was poor;

[0428] For example, protein content Moisture content amylose content The proportion of cracked grains, yellowed grains, and insect-damaged grains Adjust parameters Weighting coefficient ,

[0429]

[0430]

[0431] The quality assessment of the raw materials for rice products was qualified.

[0432] Please see Figure 2 The present invention also provides a device for predicting the quality of rice product raw materials based on partial least squares method. The system is used to execute the above-described method for predicting the quality of rice product raw materials based on partial least squares method, comprising:

[0433] The spectral data acquisition module is used to select samples from rice product raw materials using a stratified sampling method, acquire spectral data using infrared spectral imaging technology, and record characteristic wavelengths and absorbances related to protein, moisture and amylose.

[0434] The content determination module is used to determine the protein content of a sample using the Kjeldahl nitrogen determination method, the moisture content using the oven drying method, and the amylose content using the iodine colorimetric method. It also constructs a content table based on the characteristic wavelengths and absorbances associated with protein, moisture, and amylose.

[0435] The model building module is used to extract the characteristic wavelengths and absorbance of samples from the relevant content tables, and to build predictive models for protein, moisture and amylose content based on partial least squares method.

[0436] The quality prediction module is used to statistically analyze yellowed and insect-damaged grains in rice product raw materials using stratified sampling estimation method, and to calculate the protein, moisture and amylose content using prediction model, and to construct a quality assessment formula for rice product raw materials for quality prediction.

[0437] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0438] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0439] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0440] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for predicting the quality of rice product raw materials based on partial least squares method, characterized in that, The specific steps include: Step 1: Select samples from rice product raw materials using stratified sampling, collect spectral data using infrared spectroscopy imaging technology, and record the characteristic wavelengths and absorbance related to protein, moisture and amylose. Step 2: Determine the protein content of the sample using the Kjeldahl method, the moisture content using the oven drying method, and the amylose content using the iodine colorimetric method. Construct a table of relevant contents based on the characteristic wavelengths and absorbances associated with protein, moisture, and amylose. Step 3: Extract the characteristic wavelengths and absorbance of the samples from the relevant content tables, and establish a prediction model for protein, moisture and amylose content based on partial least squares method; Step 4: Use stratified sampling estimation method to count yellowed and insect-damaged grains in rice product raw materials, and use prediction model to calculate protein, moisture and amylose content, and construct rice product raw material quality assessment formula for quality prediction.

2. The method for predicting the quality of rice product raw materials based on partial least squares method according to claim 1, characterized in that: Step 1, selecting samples from rice product raw materials using stratified sampling, includes the following steps: The rice product raw materials were divided into 20 equal parts by weight. For each part of rice product raw materials, 10 samples were selected by simple random sampling and divided into 20 groups.

3. The method for predicting the quality of rice product raw materials based on partial least squares method according to claim 2, characterized in that: The acquisition of spectral data using infrared spectroscopy imaging technology, and the recording of characteristic wavelengths and absorbances related to proteins, moisture, and amylose, includes the following steps: Infrared spectral imaging technology was used to focus on 20 groups of rice product raw material samples. to Bands, recording bands absorbance ,in, and And set the characteristic wavelength of the protein as The characteristic wavelength of water is and The characteristic wavelength of amylose is .

4. The method for predicting the quality of rice product raw materials based on partial least squares method according to claim 2, characterized in that: Step 2 involves determining the protein content of the sample using the Kjeldahl method, the moisture content using the oven drying method, and the amylose content using the iodine colorimetric method, including the following steps: After grinding, each group of rice product raw material samples was divided into equal portions with a mass of [missing information]. of The protein content of the samples was determined by the Kjeldahl method, the moisture content was determined by the oven drying method, and the amylose content was determined by the iodine colorimetric method. The Kjeldahl method for determining the protein content of a sample is as follows: The mass to be placed is A sample of rice product raw materials, measured in grams, was placed in a dry Kjeldahl flask, and then... Potassium sulfate, Mixtures of copper sulfate and Continue heating the concentrated sulfuric acid until the liquid turns a clear, blue-green color, then continue heating. After the Kjeldahl flask has cooled for several hours, connect it to the Kjeldahl nitrogen distillation apparatus and add it to another conical flask. A mixture of boric acid solution and bromocresol green-methyl red indicator was used. The conical flask was placed under the condenser of the distillation apparatus, with the lower end of the condenser submerged below the surface of the boric acid solution. The solution was then added to the Kjeldahl flask. The sodium hydroxide solution was diluted until the volume of the distillate, i.e., the boric acid solution, reached [a certain value]. Afterwards, use Titration of boric acid solution with hydrochloric acid standard solution resulted in the solution changing from green to grayish-red, at which point the titration was stopped. Take another Kjeldahl flask and add... Potassium sulfate, Mixtures of copper sulfate and Continue heating the concentrated sulfuric acid until the liquid turns a clear, blue-green color, then continue heating. After the Kjeldahl flask has cooled for several hours, connect it to the Kjeldahl nitrogen distillation apparatus and add it to another conical flask. A mixture of boric acid solution and bromocresol green-methyl red indicator was used. The conical flask was placed under the condenser of the distillation apparatus, with the lower end of the condenser submerged below the surface of the boric acid solution. The solution was then added to the Kjeldahl flask. The sodium hydroxide solution was diluted until the volume of the distillate, i.e., the boric acid solution, reached [a certain value]. ,use Titrate the boric acid solution with the standard hydrochloric acid solution until the solution turns grayish-red, then stop. Record the volume of standard hydrochloric acid solution consumed, which is the volume of standard hydrochloric acid solution consumed in the blank test. Calculation of nitrogen content: In the formula, This is expressed as the nitrogen content in the sample. This is expressed as the concentration of the hydrochloric acid standard solution. This represents the volume of hydrochloric acid standard solution consumed during sample titration. This represents the volume of standard hydrochloric acid solution consumed in the blank experiment. This is expressed as the quality of the sample. Expressed as the molar mass of nitrogen; Protein content conversion: In the formula, Expressed as protein content; The method for determining moisture content using the oven drying method is as follows: Select quality as The weighing bottle will have a mass of Place the rice product raw material sample into a weighing bottle and record the total mass. Place the weighing bottle containing the rice product raw material sample into the oven and set the temperature to [temperature value missing]. The drying time is After one hour, remove the weighing bottle and record the mass of the rice product raw material sample and the weighing bottle. Calculate the moisture content: In the formula, Expressed as moisture content; The method for determining amylose content using the iodine colorimetric method is as follows: Weigh amylose standard, added After adding anhydrous ethanol, add sodium hydroxide solution Heating in a boiling water bath After cooling for minutes, transfer to In the volumetric flask, use Dilute the water to the mark, then pipette water to the mark separately. Starting with The increments are sequentially increased to... A set of volume data for the solution In a volumetric flask, dilute to volume with distilled water to a concentration of [value missing]. Starting with The increments are sequentially increased to... A series of standard solutions with a set of concentration data were prepared. Amylopectin standard was weighed and prepared according to the above method for preparing amylose standard solutions to a concentration of [missing value]. The solution was then diluted with the same gradient to prepare a series of standard solutions of different concentrations; weigh out... Dissolve potassium iodide in a small amount of distilled water, add After the iodine has completely dissolved, dilute to the final volume with distilled water. Store in a brown bottle; weigh of Sodium hydroxide solution, dissolved in distilled water, cooled and then diluted to a final volume. ; quality is Rice product raw material samples were placed Add to the conical flask Stir well with anhydrous ethanol, soak for 1 hour, filter, cool to room temperature, and transfer the solution to... In a volumetric flask, dilute to the mark with distilled water and shake well to obtain a sample solution of rice product raw materials. Then, pipette different concentrations of amylose standard solutions. At Add to each colorimetric tube Iodine reagent, diluted to volume with distilled water, shaken well and then left to stand. Minutes, with distilled water as a blank control, at wavelength... At the same time, the absorbance of each standard solution was measured by spectrophotometer. A standard curve was plotted with the amylose concentration as the abscissa and the absorbance as the ordinate, and the regression equation of amylose was obtained. draw Rice product raw material sample solution in In the colorimetric tube, iodine reagent was added and diluted to volume according to the steps for plotting the standard curve. The solution was then shaken well and allowed to stand for 15 minutes to obtain the raw material sample of the tested rice product. The absorbance of the sample solution was measured at a specific wavelength; subsequently, a blank test was performed without adding any sample, following the sample measurement procedure, and the absorbance was measured. Blank absorbance at the specified wavelength; exist At a given wavelength, the difference between the absorbance of the rice product raw material sample solution and the blank absorbance was calculated. This difference was then compared with the amylose concentration obtained from the amylose regression equation, and the dilution factor was calculated. In the formula, Expressed as a dilution factor; The formula for calculating amylose content is: In the formula, Expressed as amylose content, Expressed as amylose concentration obtained from the amylose regression equation, in units of... , The volume of the solution in the test rice product raw material sample is expressed as [volume unit]. .

5. The method for predicting the quality of rice product raw materials based on partial least squares method according to claim 4, characterized in that: In the method of determining amylose content using the iodine colorimetric method, the specific steps for deriving the regression equation for amylose include: Construct a univariate linear regression equation: In the formula, Expressed as absorbance. Expressed as amylose concentration, Represented as intercept, Expressed as slope; The absorbance of standard solutions of amylose at different concentrations was measured experimentally to obtain a series of corresponding amylose concentrations. and absorbance data points , recorded as ; Calculate the average concentration of amylose: In the formula, This is expressed as the average concentration of amylose. Calculate the average absorbance of standard solutions of amylose at different concentrations: In the formula, The absorbance is expressed as the average value of the standard amylose solutions of different concentrations. Calculate the slope of the univariate linear regression equation: Calculate the intercept of the univariate linear regression equation: 。 6. The method for predicting the quality of rice product raw materials based on partial least squares method according to claim 5, characterized in that: In step 3, the method for establishing prediction models for protein, moisture, and amylose content based on partial least squares is as follows: Based on the extracted 20 sets of data, each set containing the absorbance of 5 characteristic wavelengths, the dimension is set as follows: Absorbance data matrix: In the formula, Represented as an absorbance data matrix; Set the dimension as Dependent variable matrix: In the formula, Represented as a dependent variable matrix; Calculate the dependent variable matrix The average of columns 1, 2, and 3: In the formula, and Represented as dependent variable matrices The average of columns 1, 2, and 3. Represented as the row index of the matrix; Calculate the dependent variable matrix Standard deviations in columns 1, 2, and 3: In the formula, and Represented as a dependent variable matrix Standard deviations in columns 1, 2, and 3; For absorbance data matrix and dependent variable matrix Standardization process: In the formula, Represented as raw data, where, Represented as column indices of a matrix. , Represented as an absorbance data matrix column index, Represented as a dependent variable matrix column index, Represented as the first The average of the column, Represented as the first The standard deviation of the column is used to substitute the standardized result back into the original absorbance data matrix. and dependent variable matrix ; Among them, the The average value of the column is: No. The standard deviation of the column is: Calculate the absorbance data matrix and dependent variable matrix Covariance matrix: In the formula, Represented as an absorbance data matrix and dependent variable matrix The covariance matrix; Set the dimension as The process matrix is ​​as follows: In the formula, Represented as a process matrix; Set the dimension as The process matrix is ​​as follows: In the formula, Represented as a process two matrix; Calculate eigenvalues ​​and eigenvectors: because ,have to: In the formula, Represented as eigenvalues, Represented as an eigenvector, Represented as dimension The identity matrix; The characteristic equation of the matrix in the construction process: In the formula, The characteristic equation of the process two matrix is ​​expressed as follows: eigenvalues ; For each eigenvalue Solve the homogeneous linear equation system The corresponding feature vectors are obtained. ; contrast eigenvalues Given the size of the eigenvalues, find the largest eigenvalue and denote it as . Its corresponding eigenvector is denoted as ; Next, calculate the number of principal components to be extracted: Set the number of principal components to be For the first The variance contribution rate of each principal component is: In the formula, Represented as the first The variance contribution rate of each principal component; forward The formula for calculating the cumulative variance contribution rate of each principal component is: Starting with the first principal component, calculate the cumulative variance contribution rate sequentially. When it first reaches or exceeds 0.85, That is, the number of principal components that are determined; Calculate the absorbance data matrix First component: In the formula, Represented as an absorbance data matrix The first component; Calculate the dependent variable right Regression coefficients: In the formula, Represented as dependent variable right The regression coefficients; Calculate the dependent variable matrix First component: In the formula, Represented as a dependent variable matrix The first component; Calculate the absorbance data matrix The residual matrix: In the formula, Represented as an absorbance data matrix The residual matrix, Represented as the first load vector, the calculation formula is: Calculate the dependent variable matrix The residual matrix: In the formula, Represented as a dependent variable matrix The residual matrix; For absorbance data matrix residual matrix and dependent variable matrix residual matrix Repeat the above steps until the first one is extracted. For ingredients and : calculate Covariance matrix: In the formula, Represented as covariance matrix Calculate the first The residual matrix at the next iteration and Cross-covariance feature matrix: In the formula, Represented as the cross-covariance feature matrix; Perform eigenvalue decomposition on it, and select the eigenvector corresponding to the largest eigenvalue, denoted as . ,calculate The One component: calculate right Regression coefficients: In the formula, Represented as right The regression coefficients; calculate The One component: Calculate the first Residual matrix of the extracted components, Part 1: In the formula, Represented as the first The residual matrix of the extracted components, Represented as the first There are load vectors, where... Calculate the first Residual matrix of the second extracted component: In the formula, Represented as the first The second residual matrix of the extracted components; Constructing the component matrix: In the formula, Represented as a component matrix; Constructing the regression coefficient matrix: In the formula, Represented as a regression coefficient matrix; Construct a standardized regression model: because ,set up The standardized prediction model is as follows: Standardized prediction models Denormalized, the protein prediction model is as follows: In the formula, Represented as a predictive model for proteins; The moisture prediction model is as follows: In the formula, Represented as a moisture prediction model; In the formula, This is represented as a prediction model for amylose.

7. The method for predicting the quality of rice product raw materials based on partial least squares method according to claim 1, characterized in that: Step 4, which uses stratified sampling estimation to count yellowed and insect-damaged grains in rice product raw materials, includes the following steps: Rice product raw materials are divided into equal weight categories Groups, draw equal weight from each group Rice raw materials were used as the test samples. A Satake grain size analyzer (Japan) was used to identify cracked, yellowed, and insect-damaged grains in each sample. The identified cracked, yellowed, and insect-damaged grains were weighed separately, and their weights were recorded. ,in, ; Calculate the percentage of cracked, yellowed, and insect-damaged particles in each sample group: In the formula, Represented as the first The proportion of cracked particles, yellowed particles, and insect-damaged particles; Calculate the percentage of cracked grains, yellowed grains, and insect-damaged grains in the raw materials for rice products: In the formula, This represents the percentage of cracked grains, yellowed grains, and insect-damaged grains in the raw materials for rice products.

8. The method for predicting the quality of rice product raw materials based on partial least squares method according to claim 6, characterized in that: In step 4, the protein, moisture, and amylose content are calculated using a prediction model. Includes the following steps: for Group of samples to be tested, predict the first The method for determining the content is as follows: Represented as protein, Represented as moisture. Represented as amylose: Extract the bands respectively The absorbance, the construction dimension is The absorbance data matrix to be detected : The absorbance data matrix to be detected Standardization process: In the formula, Represented as a row index, Represented as a column index, Represented as the first element of the normalized absorbance data matrix to be detected. OK, The data in the column were used to obtain a standardized absorbance data matrix for testing. ; The computational dimension is The new sample was extracted Projection matrices along the directions of the principal components: In the formula, This is represented as a new sample in the extraction. Projection matrices along the directions of each principal component; The projection of the new sample onto the principal component direction after standardization is: This is represented as the projection of the new sample onto the principal component direction after standardization. This represents the index of the principal component, where, , Represented as The Middle Line number Elements of column index, Represented as The Middle Line number Elements of the column index; In the formula, Represented as The matrix, Represented as dimension Estimates of the regression coefficient matrix; Calculate the first The new sample in the first Standardized intermediate predicted values ​​of seed content: In the formula, Represented as the first The first sample to be tested Standardized intermediate predicted values ​​of species content, Represented as The Middle Line number Column elements; Calculate the first The nth sample to be tested was obtained after principal component projection and regression coefficient calculation. The standardized prediction formula for the content of a species is: In the formula, Represented as the first The first sample to be tested Standardized predicted values ​​of the content of each component. Represented as the first The sample to be tested was in the first... The residual in the prediction of the content of each component is set to 0. Substituting into the prediction model, then the first... The first new sample The formula for predicting the content of each component is: In the formula, Represented as the first The first new sample Predicted values ​​of component content Represented as a dependent variable matrix The Middle The standard deviation of the column, Represented as a dependent variable matrix The Middle The average of the column; The protein content of rice product raw materials is: In the formula, This refers to the protein content of rice-based raw materials; The moisture content of rice product raw materials is: In the formula, This refers to the moisture content of the raw materials used in rice products. The amylose content of rice product raw materials is: In the formula, This refers to the amylose content of rice product raw materials.

9. The method for predicting the quality of rice product raw materials based on partial least squares method according to claim 8, characterized in that: In step 4, the method for constructing a quality assessment formula for rice product raw materials to predict quality is as follows: Based on the protein, moisture, and amylose content of rice product raw materials, and the proportions of cracked, yellowed, and insect-damaged grains, a quality assessment formula for rice product raw materials is constructed: In the formula, This is expressed as a formula for evaluating the quality of rice product raw materials. and Represented as weighting coefficients, and , This is represented as an adjustment parameter; Set the quality threshold as and ,and ,when At that time, the quality assessment of rice product raw materials was excellent. At that time, the quality assessment of rice product raw materials was good. At that time, the quality assessment of rice product raw materials was qualified. At that time, the quality assessment of rice product raw materials was poor.

10. A device for predicting the quality of rice product raw materials based on partial least squares method, characterized in that: The apparatus is used to perform the method for predicting the quality of rice product raw materials based on partial least squares as described in any one of claims 1-9, comprising: The spectral data acquisition module is used to select samples from rice product raw materials using a stratified sampling method, acquire spectral data using infrared spectral imaging technology, and record characteristic wavelengths and absorbances related to protein, moisture and amylose. The content determination module is used to determine the protein content of a sample using the Kjeldahl nitrogen determination method, the moisture content using the oven drying method, and the amylose content using the iodine colorimetric method. It also constructs a content table based on the characteristic wavelengths and absorbances associated with protein, moisture, and amylose. The model building module is used to extract the characteristic wavelengths and absorbance of samples from the relevant content tables, and to build predictive models for protein, moisture and amylose content based on partial least squares method. The quality prediction module is used to statistically analyze yellowed and insect-damaged grains in rice product raw materials using stratified sampling estimation method, and to calculate the protein, moisture and amylose content using prediction model, and to construct a quality assessment formula for rice product raw materials for quality prediction.

Images