A meat quality evaluation method based on single factor variance analysis

Abstract

The application relates to a meat quality evaluation method based on single-factor variance analysis, which comprises the following steps: acquiring multi-sample data, performing standardization processing on the multi-sample data, sequentially performing oneway-ANOVA analysis and principal component analysis on the multi-sample data after the standardization processing, and generating the weight of each index in target classification; extracting the first preset number of indexes in each target classification for one-sided evaluation, obtaining individual single-item scores, taking the individual single-item scores as dependent variables to construct a multiple linear regression equation model, acquiring individual single-item score prediction values, constructing a comprehensive multiple linear regression equation model, and obtaining individual comprehensive scores. The application can quickly, accurately, conveniently and objectively evaluate pork quality.

Description

Technical Field

[0001] This invention relates to the field of livestock and poultry technology, and in particular to a method for evaluating meat quality based on one-way ANOVA. Background Technology

[0002] China is the world's largest producer and consumer of pork. In 2022, China's total hog slaughter approached 700 million head, with pork production reaching 55.41 million tons. The average per capita consumption of pork in Chinese households was 26.9 kg (data source: National Bureau of Statistics). With my country's rapid economic development, consumer demand for pork has shifted from quantity to quality; consumers are no longer simply satisfied with having meat to eat, but also with eating high-quality meat. Starting in the 1990s, the introduced Duroc-Landrace-Large White pig breed quickly captured 80% of China's pork market due to its fast growth rate and high lean meat percentage. However, Duroc-Landrace-Large White pig pork has low intramuscular fat, high tenderness, low amino acid content, and less appealing flavor. In recent years, consumers have begun to seek pork with superior quality and flavor. Brands such as "Native Pig No. 1" and "XX Black Pig" have emerged nationwide, using superior meat quality as a selling point to penetrate the premium pork market.

[0003] pH value is one of the important indicators of pork quality. It not only directly reflects muscle acidity but also has a direct impact on muscle quality. It can be used as a basis for assessing pork quality and determining the pre-slaughter health status of pigs. However, the color of pork itself does not have a significant impact on the taste of the meat; it is most directly related to consumers' purchasing desire. Muscle color mainly depends on the myoglobin content and the chemical valence state of iron atoms in the muscle. Normal meat is bright red, while protein denaturation or microbial contamination will result in a green color. Abnormal meat is grayish-white (PSE meat) or dark brown (DFD meat). Water-holding capacity refers to the ability of muscle to retain water under external force, affecting the color, aroma, taste, nutritional components, juiciness, tenderness, and other edible qualities of meat. It can be determined using the drip loss method. Marbling refers to the distribution of fat deposition in the muscle, which directly affects consumers' purchasing desire. The richer and more evenly distributed the marbling, the more tender the muscle is, and the better the chewing experience. Hardness (tenderness) usually refers to the softness of the meat and can reflect the structural characteristics of various proteins in the meat. The texture analyzer can simultaneously measure indicators such as elasticity, height, and fracturing properties of meat, characterizing its textural characteristics. Fat, protein, amino acids, vitamins, and trace elements are all nutritional indicators of meat, characterizing the product's nutritional properties. Flavor compounds mainly include amino acids and flavor nucleotides. Aspartic acid and threonine are the main umami-producing amino acids, and flavor nucleotides are mainly composed of five nucleotides: inosinic acid, guanylic acid, adenosine acid, cytidine acid, and uridine acid. The methods for measuring all indicators are shown in Table 1 below.

[0004] Table 1 Evaluation Indicators for High-Quality Pork In conclusion, the methods described above are merely measurement tools and do not definitively determine the advantages and flavor of pork. Furthermore, as people's living standards continue to improve, consumers and producers are placing higher demands on pork quality. On the one hand, they require pork to be hygienic and safe, with good taste and flavor; on the other hand, they require a vibrant red color, minimal moisture loss, and superior nutritional value. Therefore, how to quickly, accurately, conveniently, and objectively assess pork quality is an urgent problem that needs to be solved. Summary of the Invention

[0005] To address the problems existing in the prior art, the purpose of this invention is to provide a meat quality evaluation method based on one-way ANOVA, which can quickly, accurately, conveniently, and objectively assess pork quality.

[0006] To achieve the above objectives, the present invention provides the following solution: A method for evaluating meat quality based on one-way ANOVA, comprising: Acquire multi-sample data, standardize the multi-sample data, and perform oneway-ANOVA analysis and principal component analysis on the standardized multi-sample data in sequence to generate the weights of each indicator in the target classification. Extract a pre-set number of indicators from each target category for unilateral evaluation to obtain individual item scores. Use these individual item scores as dependent variables to construct a multiple linear regression equation model to obtain predicted values ​​of individual item scores. These predicted values ​​are then used to construct a comprehensive multiple linear regression equation model to obtain the overall individual score.

[0007] Optionally, the multi-sample data includes: elements, amino acids, flavor nucleotides, pH value, and data from agricultural research institutes; The elements include: calcium, iron, zinc, copper, selenium, magnesium, potassium, manganese, chromium, sodium, aluminum, nickel, lead, arsenic and cadmium; The amino acids include: ASP aspartic acid, GLU glutamic acid, SER serine, THR threonine, PRO proline, GLY glycine, ALA alanine, VAL valine, MET methionine, ILE isoleucine, LEU leucine, TYR tyrosine, PHE phenylalanine, HIS histidine, LYS lysine, and ARG arginine. The flavor nucleotides include: CMP flavor nucleotide, UMP flavor nucleotide, GMP flavor nucleotide, IMP flavor nucleotide and AMP flavor nucleotide; The data from the Academy of Agricultural Sciences includes: moisture content, protein content, water-holding capacity, electrical conductivity, flesh color, muscle fat, shear force, height, hardness, fracture properties, adhesion, elasticity, chewiness, adhesiveness, cohesion, and resilience.

[0008] Optionally, the weights of each indicator in the target classification are generated as follows: Oneway-ANOVA analysis was performed on the standardized multi-sample data to obtain indicators of interspecies differences; Principal component analysis is performed on the indicators that exhibit interspecies differences according to the target classification to generate the weights of each indicator in the target classification.

[0009] Optionally, the target classification includes: flavor, nutrition, appearance, and texture.

[0010] Optionally, obtaining the individual's individual score includes: ; in, For individual item scores, For flavor, nutrition, appearance or texture, The values ​​are standardized. This represents the weight of the indicator in the classification.

[0011] Optionally, obtaining the predicted individual score includes: ; in, This is the predicted value for an individual's single-item score. For flavor, nutrition, appearance or texture, To convert breeds into numerical representations from 1 to 7, Jiading Meishan pigs use 1, Chongming Shawutou pigs use 2, Shanghai White pigs use 3, Jinshan Fengjing pigs use 4, Pudong White pigs use 5, Songlin Topek non-green pigs use 6, and Songlin Topek green pigs use 7. The values ​​are standardized. The intercept is... The coefficient corresponding to the variety, These are the coefficients corresponding to each indicator.

[0012] Optionally, constructing the comprehensive multiple linear regression equation model includes: Based on the weights of indicators in the target categories, the multiple linear regression equation models under each target category, and the measurement difficulty of the indicator factors, multiple target indicator factors are retained to construct the comprehensive multiple linear regression equation model: ; in, For individual comprehensive scores, To convert breeds into numerical representations from 1 to 7, Jiading Meishan pigs use 1, Chongming Shawutou pigs use 2, Shanghai White pigs use 3, Jinshan Fengjing pigs use 4, Pudong White pigs use 5, Songlin Topek non-green pigs use 6, and Songlin Topek green pigs use 7. , , , , , , , , , , It consists of multiple target indicators and factors.

[0013] The beneficial effects of this invention are as follows: This invention provides a comprehensive evaluation method for pork, integrating sensory, physicochemical, nutritional, and flavor aspects, offering consumers a reference for selecting high-quality meat. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a flowchart of a meat quality evaluation method based on one-way ANOVA according to an embodiment of the present invention; Figure 2 The following are combined graphs showing the proportions of variance explained by each eigenvector in the principal component analysis of the present invention: (a) is a combined graph showing the proportions of variance explained by each eigenvector in the principal component analysis according to flavor in the target classification; (b) is a combined graph showing the proportions of variance explained by each eigenvector in the principal component analysis according to nutrition in the target classification; (c) is a combined graph showing the proportions of variance explained by each eigenvector in the principal component analysis according to appearance in the target classification; and (d) is a combined graph showing the proportions of variance explained by each eigenvector in the principal component analysis according to texture in the target classification. Figure 3 The following are individual score combination diagrams for embodiments of the present invention: (a) is an individual score diagram for flavor according to the target classification, (b) is an individual score diagram for nutrition according to the target classification, (c) is an individual score diagram for appearance according to the target classification, and (d) is an individual score diagram for texture according to the target classification. Figure 4 This is a schematic diagram of the individual comprehensive score in an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] like Figure 1 As shown, this embodiment discloses a meat quality evaluation method based on one-way ANOVA, including: acquiring multi-sample data, standardizing the multi-sample data, performing one-way ANOVA and principal component analysis on the standardized multi-sample data in sequence to generate the weights of each indicator in the target classification; extracting the first preset number of indicators in each target classification for one-way evaluation to obtain individual single-item scores, using the individual single-item scores as dependent variables to construct a multiple linear regression equation model, obtaining the predicted values ​​of individual single-item scores, which are used to construct a comprehensive multiple linear regression equation model to obtain the comprehensive individual score.

[0019] Furthermore, the multi-sample data included: elements, amino acids, flavor nucleotides, pH value, and data from the Chinese Academy of Agricultural Sciences; among which, elements included: calcium, iron, zinc, copper, selenium, magnesium, potassium, manganese, chromium, sodium, aluminum, nickel, lead, arsenic, and cadmium; amino acids included: ASP aspartic acid, GLU glutamic acid, SER serine, THR threonine, PRO proline, GLY glycine, ALA alanine, VAL valine, MET methionine, ILE isoleucine, LEU leucine, TYR tyrosine, PHE phenylalanine, HIS histidine, LYS lysine, and ARG arginine; flavor nucleotides included: CMP flavor nucleotides, UMP flavor nucleotides, GMP flavor nucleotides, IMP flavor nucleotides, and AMP flavor nucleotides; data from the Chinese Academy of Agricultural Sciences included: moisture content, protein content, water-holding capacity, electrical conductivity, meat color, muscle fat, shear force, height, hardness, fracture properties, adhesion, elasticity, chewiness, adhesiveness, cohesion, and resilience.

[0020] Furthermore, generating the weights of each indicator in the target classification includes: performing one-way ANOVA analysis on the standardized multi-sample data to obtain indicators with interspecific differences; and performing principal component analysis on the indicators with interspecific differences according to the target classification to generate the weights of each indicator in the target classification. More specifically: S1. In this invention, multi-sample data is first standardized using the following formula: ; in, These are standardized values. It is a variable. It is an expectation. It is the standard deviation.

[0021] S2. One-way ANOVA analysis was performed on the standardized multi-sample data. This invention found that only 44 out of 53 indicators showed interspecific differences, and these 44 indicators were retained. Principal component analysis was then performed on the filtered standardized multi-sample data for flavor, nutrition, appearance, and texture, respectively. The first principal component analysis explained 49.4%, 68.2%, 52.6%, and 42.2% of the variance, respectively. Figure 2 As shown in (a)-(d), the results yielded the weights of each indicator in the classification. AMP and GMP had the highest weights in flavor, at 65.67 and 61.26 respectively, as shown in Table 2. In the nutritional indicator evaluation, as shown in Table 3, in addition to aluminum (96.95), fat content was also high, with a weight of 15.63. In the appearance evaluation indicators, as shown in Table 4, the top three weights were a, muscle fat, and L, with weights of 82.30, 44.32, and 25.13 respectively. In the texture indicator evaluation, as shown in Table 5, five indicators had weights exceeding 20: adhesion, shear strength, fracture strength, hardness, and chewiness, with weights ranging from 75.22 to 23.34.

[0022] Table 2. Weighting of Sub-indicators in Flavor Table 3. Weighting of Sub-indicators in Nutrition Table 4. Weighting of Sub-indicators in Appearance Table 5. Weighting of Sub-indicators in Texture S3. Select the top 30% of indicators in each category and construct individual evaluation indicators for each. The construction method is as follows: multiply the weights of the top 30% of indicators by their standardized values ​​and then add them together to obtain the individual score, i.e., the evaluation formulas for the individual indicators of flavor, nutrition, appearance, and texture: ; in, It is an individual's score for a single item. Is it flavor, nutrition, appearance, or texture? These are standardized values. It is the weight of the indicator in the classification.

[0023] S4. Subsequently, this invention studies the dependent variable using statistical methods. With independent variable , … The relationship between them can be explained by a multiple linear regression equation. The multiple linear regression equation model is shown below: ; in It is the predicted value of an individual's single-item score. Is it flavor, nutrition, appearance, or texture? The breeds are represented by numbers 1-8 (1 for "Jiading Meishan Pig", 2 for "Chongming Shawutou", 3 for "Shanghai White Pig", 4 for "Jinshan Fengjing Pig", 5 for "Pudong White Pig", 6 for "Songlin Topek Non-Green Pig", and 7 for "Songlin Topek Green Label Pig"). These are standardized values. It is the intercept. It is the coefficient corresponding to the variety. These are the coefficients corresponding to each indicator.

[0024] In the study of this invention, S5 obtained individual scores for 82 individuals under four categories, as shown in the following results. Figure 3 As shown in (a)-(d).

[0025] S6 uses a multiple regression model to infer the relationship between an individual's flavor score and variety, AMP, and GMP: ; Using the same method as S7, this invention presumes the relationship between an individual's nutritional score and breed, aluminum, fat content, selenium, copper, calcium, and proline: ; S8 This invention presumes the relationship between an individual's appearance score and breed, a, and intramuscular fat as follows: ; S9 This invention presumes the relationship between an individual's texture score and its adhesion and shear strength as follows: ; S10 comprehensively considers the weight of indicators in the classification, the multiple linear regression equation model under each classification, and the measurement difficulty of factors. This invention retains 10 factors: AMP, GMP, iron, moisture content, red-green hue (a), yellow-blue hue (b), color brightness (L), muscle fat, shear force, and hardness. Therefore, this invention can obtain an individual comprehensive score, such as... Figure 4As shown, and based on this, a comprehensive multiple linear regression equation model is constructed as follows: ; in, For individual comprehensive scores, To convert breeds into numerical representations from 1 to 7, Jiading Meishan pigs use 1, Chongming Shawutou pigs use 2, Shanghai White pigs use 3, Jinshan Fengjing pigs use 4, Pudong White pigs use 5, Songlin Topek non-green pigs use 6, and Songlin Topek green pigs use 7. , , , , , , , , , , It consists of multiple target indicators and factors.

[0026] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for evaluating meat quality based on one-way ANOVA, characterized in that, include: Acquire multi-sample data, standardize the multi-sample data, and perform oneway-ANOVA analysis and principal component analysis on the standardized multi-sample data in sequence to generate the weights of each indicator in the target classification. Extract a pre-set number of indicators from each target category for unilateral evaluation to obtain individual item scores. Use these individual item scores as dependent variables to construct a multiple linear regression equation model to obtain predicted values ​​of individual item scores. These predicted values ​​are then used to construct a comprehensive multiple linear regression equation model to obtain the overall individual score.

2. The meat quality evaluation method based on one-way ANOVA according to claim 1, characterized in that, The multi-sample data includes: elements, amino acids, flavor nucleotides, pH value, and data from the Academy of Agricultural Sciences; The elements include: calcium, iron, zinc, copper, selenium, magnesium, potassium, manganese, chromium, sodium, aluminum, nickel, lead, arsenic and cadmium; The amino acids include: ASP aspartic acid, GLU glutamic acid, SER serine, THR threonine, PRO proline, GLY glycine, ALA alanine, VAL valine, MET methionine, ILE isoleucine, LEU leucine, TYR tyrosine, PHE phenylalanine, HIS histidine, LYS lysine, and ARG arginine. The flavor nucleotides include: CMP flavor nucleotide, UMP flavor nucleotide, GMP flavor nucleotide, IMP flavor nucleotide and AMP flavor nucleotide; The data from the Academy of Agricultural Sciences includes: moisture content, protein content, water-holding capacity, electrical conductivity, flesh color, muscle fat, shear force, height, hardness, fracture properties, adhesion, elasticity, chewiness, adhesiveness, cohesion, and resilience.

3. The meat quality evaluation method based on one-way ANOVA according to claim 1, characterized in that, The weights of each indicator in the target classification are generated as follows: Oneway-ANOVA analysis was performed on the standardized multi-sample data to obtain indicators of interspecies differences; Principal component analysis is performed on the indicators that exhibit interspecies differences according to the target classification to generate the weights of each indicator in the target classification.

4. The meat quality evaluation method based on one-way ANOVA according to claim 3, characterized in that, The target categories include: flavor, nutrition, appearance, and texture.

5. The meat quality evaluation method based on one-way ANOVA according to claim 1, characterized in that, The individual scores obtained include: ； in, For individual item scores, For flavor, nutrition, appearance or texture, The values ​​are standardized. This represents the weight of the indicator in the classification.

6. The meat quality evaluation method based on one-way ANOVA according to claim 1, characterized in that, Obtaining the predicted individual score includes: ； in, This is the predicted value for an individual's single-item score. For flavor, nutrition, appearance or texture, To convert breeds into numerical representations from 1 to 7, Jiading Meishan pigs use 1, Chongming Shawutou pigs use 2, Shanghai White pigs use 3, Jinshan Fengjing pigs use 4, Pudong White pigs use 5, Songlin Topek non-green pigs use 6, and Songlin Topek green pigs use 7. The values ​​are standardized. The intercept is... The coefficient corresponding to the variety, These are the coefficients corresponding to each indicator.

7. The meat quality evaluation method based on one-way ANOVA according to claim 1, characterized in that, Constructing the comprehensive multiple linear regression equation model includes: Based on the weights of indicators in the target categories, the multiple linear regression equation models under each target category, and the measurement difficulty of the indicator factors, multiple target indicator factors are retained to construct the comprehensive multiple linear regression equation model: ； in, For individual comprehensive scores, To convert breeds into numerical representations from 1 to 7, Jiading Meishan pigs use 1, Chongming Shawutou pigs use 2, Shanghai White pigs use 3, Jinshan Fengjing pigs use 4, Pudong White pigs use 5, Songlin Topek non-green pigs use 6, and Songlin Topek green pigs use 7. , , , , , , , , , , It consists of multiple target indicators and factors.

Images