Hami melon three-dimensional quality evaluation method based on multi-source information fusion

Abstract

The invention belongs to the technical field of fresh fruit quality detection, and provides a Hami melon three-dimensional quality evaluation method based on multi-source information fusion, and the method comprises the following steps: collecting and obtaining multi-source information data of a to-be-detected Hami melon sample; constructing a sensory sweetness dimension evaluation model, and obtaining a sensory sweetness value according to the multi-source information data; constructing a sensory brittleness dimension evaluation model, and obtaining a sensory brittleness value according to the multi-source information data; constructing a sensory aroma dimension evaluation model, and obtaining a sensory aroma value according to the multi-source information data; and constructing a three-dimensional quality evaluation model, and obtaining a comprehensive quality score of the Hami melon sample according to the sensory sweetness value, the sensory brittleness value and the sensory aroma value. According to the Hami melon three-dimensional quality evaluation method based on multi-source information fusion provided by the invention, quantitative evaluation of three dimensions of sweetness, brittleness and fragrance of the Hami melons is integrated, so that the evaluation result is closer to the real comprehensive quality of the Hami melons, and the objectivity and accuracy of the evaluation result are improved.

Description

A Three-Dimensional Quality Evaluation Method for Hami Melon Based on Multi-Source Information Fusion Technical Field

[0001] This invention belongs to the field of fresh fruit quality testing technology, and specifically relates to a three-dimensional quality evaluation method for Hami melon based on multi-source information fusion. Background Technology

[0002] Hami melon is a general term for a group of thick-skinned melon varieties native to Xinjiang. Its fruit is sweet, crisp, and nutritious, and possesses medicinal properties, making it highly popular among consumers and earning it the title of "China's No. 1 Honeydew Melon." It is widely distributed in Xinjiang, Gansu, Henan, Shandong, Hainan, and other regions, with an annual production of approximately 16 million tons. As people's living standards continue to improve, consumers are increasingly emphasizing the internal and external quality of Hami melons. However, current methods for evaluating the quality of Hami melons still largely rely on manual sensory assessment. While this can broadly reflect the overall quality of the melon, it suffers from limitations such as strong subjectivity, susceptibility to human judgment, low efficiency, and a lack of standardized criteria. These shortcomings result in unreliable and inaccurate assessments, hindering widespread and large-scale adoption. Furthermore, although some existing technologies use specialized instruments such as liquid chromatography, saccharimeters, or gas chromatography-mass spectrometry to detect quality parameters of fresh fruit and quantify the detection indicators, and evaluate the sweetness or ripeness of fresh fruit based on the detected quantitative indicators, they often focus on the detection and evaluation of a single indicator. For example, they use a saccharimeter to detect the sugar content and evaluate the sweetness of fresh fruit, or use gas chromatography-mass spectrometry (GC-MS) to analyze aroma compounds and evaluate the ripeness of fresh fruit. Obviously, the above evaluation methods can only evaluate a single characteristic of fresh fruit and cannot reflect the overall comprehensive quality of fresh fruit.

[0003] Therefore, it is necessary to design a three-dimensional quality evaluation method for Hami melons based on multi-source information fusion that can at least solve some of the above problems and defects. Summary of the Invention

[0004] To address the above technical problems, this invention proposes a three-dimensional quality evaluation method for Hami melons based on multi-source information fusion. This method integrates quantitative evaluations of three dimensions: sweetness, crispness, and aroma, making the evaluation results closer to the true overall quality of Hami melons, thereby improving the objectivity and accuracy of the evaluation results.

[0005] The technical solution of this invention is:

[0006] This invention proposes a three-dimensional quality evaluation method for Hami melons based on multi-source information fusion, comprising the following steps:

[0007] S1, Collect and acquire multi-source information data of the cantaloupe sample to be tested;

[0008] S2, construct a sensory sweetness dimension evaluation model, and obtain the sensory sweetness value of the cantaloupe sample to be tested based on the multi-source information data;

[0009] S3, construct a sensory crispness dimension evaluation model, and obtain the sensory crispness value of the cantaloupe sample to be tested based on the multi-source information data;

[0010] S4, construct a sensory aroma dimension evaluation model, and obtain the sensory aroma value of the cantaloupe sample to be tested based on the multi-source information data;

[0011] S5. Construct a three-dimensional quality evaluation model and obtain the comprehensive quality score of the cantaloupe sample based on the sensory sweetness value, the sensory crispness value, and the sensory aroma value.

[0012] Preferably, the multi-source information data includes the glucose content, sucrose content, fructose content, flesh firmness, flesh crispness, flesh compactness, and content of key aroma active ingredients of the cantaloupe sample to be tested.

[0013] Preferably, the sensory sweetness dimension evaluation model includes a sensory sweetness value conversion formula:

[0014] ;

[0015] In the formula: The sensory sweetness value. For glucose content, This is the weighting factor for the sweetness of glucose. For sucrose content, This is the sweetness weighting coefficient for sucrose. For fructose content, This is the sweetness weighting coefficient for fructose, and The value is 0.75. The value is 1.0. The value is 1.7.

[0016] Preferably, the sensory crispness dimension evaluation model includes a sensory crispness value conversion formula:

[0017] ;

[0018] In the formula: Sensory crispness value, The firmness of the fruit flesh, This is a weighting coefficient for the crispness of the fruit flesh. The flesh is crisp. This is the weighting coefficient for the crispness of the fruit flesh. For the firmness of the fruit flesh, The crispness weighting coefficient represents the firmness of the fruit flesh. For the offset constant term, and The value is -0.00114. The value is 0.0008146. The value is -0.00146. The value is 2.7948.

[0019] Preferably, the sensory aroma dimension evaluation model includes a sensory aroma value conversion formula:

[0020] ;

[0021] In the formula: The sensory aroma value, The first among the key aroma active ingredients The content of each component, The first among the key aroma active ingredients Aroma weight coefficients corresponding to each component.

[0022] Preferably, the key aroma active ingredient is obtained through the following steps:

[0023] Volatile components of cantaloupe aroma compounds were collected, the odor activity value of each volatile component was calculated, and volatile components with odor activity values ​​greater than or equal to 1 were screened to obtain the key aroma active components.

[0024] The odor activity value is calculated as follows:

[0025] ;

[0026] In the formula: For the first The odor activity value of the volatile components. For the first The content of volatile components, For the first Odor threshold of a certain volatile component in water.

[0027] Preferably, the key aroma active ingredients include (Z)-2-nonenal, ethyl acetate, trans-2-,cis-6-nonadienal, methyl 2-methylbutyrate, 1-octen-3-ol, and ethyl 2-methylbutyrate.

[0028] Preferably, the aroma weight coefficient corresponding to the key aroma active ingredient is calculated through the following steps:

[0029] Based on the hierarchical analysis, a hierarchical structure was constructed, including a target layer with the evaluation objective of excellent Hami melon aroma quality and a constraint layer with key aroma active components as evaluation criteria.

[0030] The importance of key aroma active ingredients is compared pairwise using the scaling method, thereby constructing a judgment matrix. The resulting judgment matrix is ​​then subjected to a consistency test to ensure that it meets the consistency test criteria.

[0031] The weight column vector is obtained by calculating the judgment matrix, and finally the aroma weight coefficient corresponding to each key aroma active ingredient is obtained.

[0032] Preferably, the three-dimensional quality evaluation model includes a comprehensive quality score conversion formula:

[0033] ;

[0034] In the formula: To determine the overall quality score, The sensory sweetness value. The weighting coefficients corresponding to the sensory sweetness values. Sensory crispness value, The weighting coefficient corresponding to the sensory crispness value. The sensory aroma value, The weighting coefficients corresponding to the sensory aroma values. and .

[0035] Preferably, the three-dimensional quality evaluation method for Hami melon based on multi-source information fusion provided by the present invention further includes the following steps:

[0036] S6. A grading standard is constructed based on the comprehensive quality score of the cantaloupe samples. The grading standard includes special grade, first grade and second grade.

[0037] The present invention has the following advantages and effects compared with the prior art:

[0038] (1) A three-dimensional quality evaluation model is adopted to obtain a comprehensive quality score through sensory sweetness value, sensory crispness value and sensory aroma value. It integrates and consolidates the quantitative evaluation sub-models including sweetness, crispness and aroma, thus forming a three-dimensional quality evaluation model with multiple dimensions. It can comprehensively and objectively evaluate the quality of Hami melon. Compared with the existing single sugar index or hardness index evaluation standards, it can more comprehensively reflect the overall quality and commercial value of Hami melon. The evaluation results are objective, accurate and practical.

[0039] (2) Using multi-source information data including glucose content, sucrose content, fructose content, flesh firmness, flesh crispness, flesh compactness, and key aroma active ingredient content of cantaloupe samples, the instrument detection replaces the existing evaluation method that relies on manual tasting, which greatly reduces the subjectivity and individual differences of the evaluation results, making the quality evaluation results more objective and accurate, and reducing subjective bias.

[0040] (3) The grading standard constructed based on the comprehensive quality score of Hami melon samples can directly and clearly output the quality score and quality grade, providing a clear and unified data guide and basis for Hami melon market grading, premium pricing, branded sales and planting guidance, which helps to improve the standardization level of the industry. Attached Figure Description

[0041] Figure 1 is a flowchart illustrating the three-dimensional quality evaluation method for Hami melon based on multi-source information fusion in an embodiment of the present invention.

[0042] Figure 2 is a clustering heatmap showing the correlation between sensory crispness value and flesh hardness, flesh crispness, and flesh firmness in the embodiments of the present invention.

[0043] Figure 3 shows the verification fitting curve of the actual sensory score of crispness and the sensory crispness value in the embodiment of the present invention;

[0044] Figure 4 is a matching network diagram of the sensory fragility dimension evaluation model in the test verification of the present invention, showing the model evaluation results (model predicted values) and the actual sensory evaluation results (sensory evaluation values).

[0045] Figure 5 shows the verification fitting curve between the aroma sensory score and the sensory aroma value in the embodiment of the present invention.

[0046] Figure 6 shows the verification fitting curve of the sensory evaluation total score and the comprehensive quality score in the embodiment of the present invention. Detailed Implementation

[0047] To enable those skilled in the art to better understand the present invention, specific embodiments will now be described in further detail. It should be understood that the specific embodiments described herein are for illustrative and explanatory purposes only and are not intended to limit the scope of the invention.

[0048] Example:

[0049] As shown in Figure 1, this invention provides a three-dimensional quality evaluation method for Hami melons based on multi-source information fusion, which specifically includes the following steps:

[0050] S1. Collect multi-source information data of the cantaloupe sample to be tested. The multi-source information data includes the glucose content, sucrose content, fructose content, flesh firmness, flesh crispness, flesh compactness, and key aroma active ingredient content of the cantaloupe sample to be tested. It should be noted that the above multi-source information data are all obtained by measuring and collecting the flesh of the cantaloupe sample to be tested. At the same time, the contents of glucose, fructose, sucrose, and organic acids can be obtained by liquid chromatography analysis, the flesh firmness, flesh crispness, and flesh compactness can be obtained by texture analysis, and the contents of volatile compounds and key aroma active ingredients can be obtained by aroma gas chromatography-mass spectrometry analysis.

[0051] Furthermore, since the physical meaning and dimensions of multi-source information data are completely different, it is necessary to perform dimensionless processing on all multi-source information data acquired through the original instruments, that is, to normalize the multi-source information data. Specifically, in this embodiment, the range transformation method is used to perform dimensionless normalization processing on the original multi-source information data acquired through the instruments:

[0052] ;

[0053] In the formula: The first in the multi-source information data after normalization processing Data such as glucose content or fruit firmness. The first sample collected by the instrument The original values ​​in the class data, The first of the overall samples collected by the instrument The minimum value in the class of data. The first of the overall samples collected by the instrument The maximum value in the class of data.

[0054] S2. A sensory sweetness evaluation model was constructed, and the sensory sweetness values ​​of the cantaloupe samples were obtained based on multi-source information data. Sugar is not only an important component of fruit quality and flavor, but also a basic raw material for the synthesis of nutrients such as amino acids and vitamin C. Fructose, glucose, and sucrose are the main soluble sugar components in cantaloupe fruits, and their content and composition are key to their quality. The accumulation of sucrose is the key to determining fruit quality. Sweetness is usually measured using sucrose as a reference value, with sucrose as 100. Fructose is almost twice as sweet, and other natural sugars are all less than sucrose. With a sweetness value of 1.00 for sucrose, the corresponding sweetness value for fructose is 1.7, and the sweetness value for glucose is 0.75.

[0055] Specifically, the sensory sweetness dimension evaluation model includes the sensory sweetness value conversion formula:

[0056] ;

[0057] In the formula: The sensory sweetness value. For glucose content, This is the weighting factor for the sweetness of glucose. For sucrose content, This is the sweetness weighting coefficient for sucrose. For fructose content, is the sweetness weighting coefficient for fructose, where, The value is 0.75. The value is 1.0. The value is 1.7.

[0058] S3. A sensory crispness evaluation model was constructed, and the sensory crispness values ​​of the cantaloupe samples were obtained based on multi-source information data. Specifically, the sensory crispness evaluation model used a sensory evaluation method to score the crispness of the cantaloupe samples and obtain sensory crispness values. Simultaneously, a texture analyzer was used to measure its textural characteristic parameters (e.g., flesh firmness, flesh crispness, flesh compactness, etc.), and the physicochemical indicators related to the crispness score and textural characteristic parameters were measured. To clarify the relationship between each indicator parameter and the sensory crispness value score, canonical correlation analysis was used to systematically examine the intrinsic correlation between textural characteristic parameters and physicochemical indicators, as well as their correlation with the sensory crispness score. Specific results are shown in Figure 2. Correlation clustering heatmap analysis revealed a significant correlation between flesh firmness, flesh crispness, flesh compactness, and the sensory crispness value. Therefore, these parameters were used as key independent variables in subsequent modeling.

[0059] Furthermore, based on the selected indicators of flesh firmness, flesh crispness, and flesh compactness, partial least squares regression analysis (PLSR) was used to construct a multiple linear regression model with the above indicators as independent variables and sensory crispness as the dependent variable. This model is the sensory crispness dimension evaluation model, which specifically includes the sensory crispness value conversion formula:

[0060] ;

[0061] In the formula: Sensory crispness value, The firmness of the fruit flesh, This is a weighting coefficient for the crispness of the fruit flesh. The flesh is crisp. This is the weighting coefficient for the crispness of the fruit flesh. For the firmness of the fruit flesh, The crispness weighting coefficient represents the firmness of the fruit flesh. For the offset constant term, where, The value is -0.00114. The value is 0.0008146. The value is -0.00146. The value is 2.7948. Referring to Figure 3, the goodness of fit (R²) of the sensory fragility dimension evaluation model is 0.9035, indicating that the model has good explanatory power and passes the statistical significance test.

[0062] To verify the applicability of the aforementioned sensory crispness evaluation model, Hami melon samples were collected from Xinjiang, Gansu, Hainan, and Shandong. Corresponding parameters such as flesh firmness, flesh crispness, and flesh compactness were measured. Ten trained sensory evaluators conducted a sensory evaluation of crispness. The measured parameters were then input into the sensory crispness evaluation model to calculate sensory crispness values, which were compared with the actual sensory scores obtained by the evaluators. Figure 3 shows the sensory crispness values ​​obtained through the sensory crispness evaluation model on the horizontal axis and the actual sensory scores obtained by the evaluators on the vertical axis. As shown in Table 1 and Figure 3, the model evaluation results and the actual sensory evaluation results generally show a consistent trend, indicating that the model has a certain degree of reliability and universality. It is understandable that, given the limited sample source and quantity of the above verification data, the stability and applicability of the model can be further verified and optimized by expanding the sample size and covering more varieties and production areas.

[0063] Table 1. Test Validation Data of Hami Melon Samples from Different Origins Variety Origin Actual Sensory Score (Sensory Evaluator) Flesh Firmness (g) Flesh Crispness (g.sec) Flesh Tightness (g.sec) Sensory Crispness Value (Model Prediction) Golden Honey Hainan 4.83 25 37 17 57 64.61 Duzhi-25 Hainan 4.54 65 39 13 67 04.47 Xizhou Honey 25 Shandong 3.84 70 38 08 86 94.09 Duzhi-25 Gansu 4.27 54 34 24 34 04.23 Xizhou Honey 17 Xinjiang 4.75 63 38 76 52 64.54 surface

[0064] S4. Construct a sensory aroma dimension evaluation model and obtain the sensory aroma value of the cantaloupe sample to be tested based on multi-source information data.

[0065] Specifically, the sensory aroma dimension evaluation model includes the sensory aroma value conversion formula:

[0066] ;

[0067] In the formula: The sensory aroma value, The first among the key aroma active ingredients The content of each component, The first among the key aroma active ingredients Aroma weight coefficients corresponding to each component.

[0068] The acquisition of key aroma active ingredients includes the following steps:

[0069] Volatile components of cantaloupe aroma compounds were collected and obtained. The odor activity value of each volatile component was calculated, which represents the intensity of the role of the aroma substances in the aroma. Volatile components with an odor activity value greater than or equal to 1 were screened to obtain the key aroma active components.

[0070] The odor activity value is calculated as follows:

[0071] ;

[0072] In the formula: For the first The odor activity value of the volatile components. For the first The content of various volatile components, expressed in mg / kg. For the first Odor threshold of a certain volatile component in water, expressed in mg / kg.

[0073] The volatile components of Hami melon aroma compounds are mainly composed of esters, aldehydes, and alcohols, supplemented by ketones and terpenes. The corresponding key aroma active components are finally obtained through the above-mentioned key aroma active component acquisition steps. The key aroma active components specifically include (Z)-2-nonenal, ethyl acetate, trans-2-,cis-6-nonadienal, methyl 2-methylbutyrate, 1-octen-3-ol, and ethyl 2-methylbutyrate.

[0074] Furthermore, the aroma weight coefficients corresponding to the key aroma active ingredients are calculated through the following steps:

[0075] Based on the Analytic Hierarchy Process (AHP), a hierarchical structure was constructed, including a target layer with the evaluation objective of superior cantaloupe aroma quality and a constraint layer with key aroma active components as evaluation criteria.

[0076] The importance of key aroma active ingredients was compared pairwise using the scaling method, and a judgment matrix was constructed accordingly. A consistency test was then performed on the resulting judgment matrix to ensure it met the consistency test criteria. Specifically, in practical applications, 10-15 industry experts (in the fields of fruit and vegetable quality research, sensory evaluation, and cantaloupe breeding) were invited to compare the importance of indicators at the same level using the 1-9 scaling method (1 = equally important, 9 = extremely important). Combining the hierarchical relationships between factors and expert opinions, the scaling method in the analytic hierarchy process (see Table 2) was used to construct a judgment matrix and perform a consistency test. The standard was a random consistency ratio (CR) < 0.1 for the judgment matrix. If the CR of each matrix was within 0.1, and the overall consistency test CR < 0.1, it indicated that the weight assignment of each factor met the requirements, and all matrices were reasonably consistent.

[0077] The weight column vector is obtained by calculating the judgment matrix, and finally the aroma weight coefficient corresponding to each key aroma active ingredient is obtained. As shown in Table 3, the weight calculation results of the six evaluation indicators are compiled graphically using the AHP method, which yields the weight distribution corresponding to the six key aroma active ingredients. The CR of the weight of the cantaloupe aroma index is 0.049, which is less than 0.1. Therefore, it can be seen that the assignment of the judgment matrix is ​​scientific and reasonable, and the obtained matrix weight values ​​are reliable.

[0078] Table 2. Scaling Method: Scale Meaning 1 Equally Important 3 Slightly (Not) Important 5 Relatively (Not) Important 7 Very (Not) Important 9 Absolutely (Not) Important 2, 4, 6, 8 Between the above categorized values surface

[0079] Table 3. Judgment Matrix (Z) for the Construction of Key Aroma Active Components - 2-Nonenal, Ethyl Acetate, Trans-2-,cis-6-Nonadienal, Methyl 2-Methylbutyrate, 1-Octen-3-ol, Ethyl 2-Methylbutyrate Feature Vector Weights (Z) - 2-Nonenal 151595 3.225 0.35 Ethyl Acetate 0.21 0.2151 0.765 0.08 Trans-2-,cis-6-Nonadienal 151777 3.46 0.38 Methyl 2-Methylbutyrate 0.21 0.143151 0.723 0.08 1-Octen-3-ol 0.111 0.2 0.143 0.21 0.2 0.224 0.02 Ethyl 2-Methylbutyrate 0.21 0.143151 0.723 0.08 surface

[0080] Note: λmax = 6.305, consistency ratio CR = 0.049 < 0.1

[0081] Combining the above steps for obtaining key aroma active ingredients and calculating aroma weighting coefficients, the final sensory aroma value conversion formula is as follows:

[0082] ;

[0083] In the formula: The sensory aroma value, The content of (Z)-2-nonenal, The aroma weighting coefficient corresponding to (Z)-2-nonenal. This refers to the content of ethyl acetate. This represents the aroma weighting coefficient corresponding to ethyl acetate. The content of trans-2-,cis-6-nonadienal, The aroma weighting coefficient for trans-2-,cis-6-nonadienal. The content of methyl 2-methylbutyrate, This represents the aroma weighting coefficient corresponding to methyl 2-methylbutyrate. The content of 1-octen-3-ol, The aroma weighting coefficient corresponding to 1-octen-3-ol. The content of ethyl 2-methylbutyrate. This represents the aroma weighting coefficient corresponding to ethyl 2-methylbutyrate.

[0084] Specifically, The value is 0.35. The value is 0.08. The value is 0.38. The value is 0.08. The value is -0.02. The value is 0.08. It should be noted that since 1-octen-3-ol has a mushroom or earthy flavor, it has a negative effect on the overall flavor of cantaloupe. Therefore, its corresponding weighting coefficient is negative.

[0085] It is understandable that after obtaining the content data of key aroma active ingredients, it is necessary to standardize them (such as using SPSS software for Z-score standardization) to ensure the consistency and representativeness of the data processing. The sensory aroma value data obtained based on the above sensory aroma dimension evaluation model and sensory aroma value conversion formula can be referred to Table 4 below.

[0086] Table 4. Content data of key aroma active ingredients and sensory aroma value data. Sample No. Ethyl acetate 2-methylbutyrate ethyl 2-methylbutyrate 1-octen-3-ol trans-2-,cis-6-nonadienal (Z)-2-nonenal Sensory aroma value S1 0.0000 0.1312 0.0000 0.3483 0.9091 1.0000 0.699 S2 0.1925 0.0293 0.0300 0.1384 0.0000 1.0000 0.367 S3 0.0709 0.0000 0.0076 0.0581 1.0000 0.7477 0.647 S4 1.0000 0.0000 0.0467 0.6127 0.0000 0.0000 0.071 surface

[0087] To verify the accuracy of the above sensory aroma dimension evaluation model, a sensory evaluation team (10 people) with systematic training was formed. Using quantitative descriptive analysis (QDA), the sensory attributes that consumers are sensitive to (such as sweet aroma, honey aroma, fruity aroma, green aroma, off-flavor, aroma intensity, aroma persistence, and overall aroma) were screened out. Finally, the overall aroma was determined as the final aroma sensory index. Sensory evaluations were conducted on the aroma of cantaloupes from different origins and varieties. The specific results are shown in Table 5 below.

[0088] Table 5. Sensory evaluation results of the aroma of Hami melon samples from different origins. (Table showing aroma scores for each type of Hami melon) surface

[0089] As can be seen from Tables 4 and 5, the sensory aroma values ​​obtained by different samples through the sensory aroma dimension evaluation model are consistent with the order of aroma scores in the sensory scoring results.

[0090] To further verify the reliability of the above sensory aroma dimension evaluation model, a linear curve was plotted with the aroma sensory score (aroma score) as the ordinate and the sensory aroma value obtained through the sensory aroma dimension evaluation model as the abscissa. The resulting linear curve is: y = 2.2987x + 3.0748. As shown in Figure 5, the fitting coefficient is relatively high, R² = 0.9908, which is close to 1.

[0091] S5. Construct a three-dimensional quality evaluation model to obtain the comprehensive quality score of the cantaloupe sample based on sensory sweetness value, sensory crispness value, and sensory aroma value.

[0092] Specifically, the three-dimensional quality evaluation model includes the comprehensive quality score conversion formula:

[0093] ;

[0094] In the formula: To determine the overall quality score, The sensory sweetness value. The weighting coefficients corresponding to the sensory sweetness values. Sensory crispness value, The weighting coefficient corresponding to the sensory crispness value. The sensory aroma value, Here, the weighting coefficients corresponding to the sensory aroma values ​​are: ,and This ensures that the weight of sweetness, crispness, and aroma in the overall quality score decreases sequentially. Specifically, in this embodiment, The value is 0.38. The value is 0.33. The value is 0.29. It should be noted that in practical applications, the above weighting coefficient can be adjusted within a range of ±10% to improve the adaptability and scalability for evaluating the quality of different Hami melons.

[0095] To verify the accuracy of the aforementioned three-dimensional quality evaluation model, cross-validation was employed. Specifically, regression analysis was used to fit the total sensory evaluation score obtained by the sensory evaluation team with the comprehensive quality score obtained by the three-dimensional quality evaluation model. The results are shown in Figure 6, where the horizontal axis represents the total sensory evaluation score and the vertical axis represents the comprehensive quality score. The correlation coefficient is 0.869, indicating that the three-dimensional quality evaluation model established using sweetness, crispness, and aroma as evaluation indicators is suitable for the comprehensive evaluation of Hami melon fruit quality, possesses high reliability, and can replace manual tasting and scoring, thereby achieving rapid and objective evaluation of Hami melon quality.

[0096] Furthermore, the three-dimensional quality evaluation method for Hami melons based on multi-source information fusion provided in this embodiment also includes the following steps:

[0097] Step S6: Construct a grading standard based on the comprehensive quality score of the cantaloupe samples. The grading standard specifically includes three levels: Special Grade, First Grade, and Second Grade. Specifically, referring to Figure 6, the grading points are determined at 0.850, 0.70, and 0.55 for the comprehensive quality score (vertical axis), dividing the cantaloupe varieties into three levels: Special Grade corresponds to a comprehensive quality score of no less than 0.85, indicating excellent cantaloupe quality with high sugar content, crisp texture, and rich aroma. It is recommended for market positioning as a core production area landmark product and high-end gift boxes. First Grade corresponds to a comprehensive quality score between 0.70 and 0.85, indicating excellent cantaloupe quality with a balanced sweetness and crispness, typical cantaloupe aroma, and no obvious defects. It is recommended for market positioning in brand supermarkets and e-commerce platforms. Second Grade corresponds to a comprehensive quality score between 0.55 and 0.70, indicating acceptable cantaloupe quality, but with slight deficiencies in sweetness, crispness, or aroma. It is recommended for market positioning in farmers' markets and mass consumption.

[0098] In summary, the three-dimensional quality evaluation method for Hami melon based on multi-source information fusion provided by this invention integrates quantitative evaluations of three dimensions: sweetness, crispness, and aroma of Hami melon. This makes the evaluation results closer to the true comprehensive quality of Hami melon, thereby improving the objectivity and accuracy of the evaluation results.

[0099] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. All equivalent changes and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A three-dimensional quality evaluation method for Hami melons based on multi-source information fusion, characterized in that, Includes the following steps: S1. Collect multi-source information data of the cantaloupe sample to be tested; S2. Construct a sensory sweetness dimension evaluation model and obtain the sensory sweetness value of the cantaloupe sample to be tested based on the multi-source information data; S3. Construct a sensory crispness dimension evaluation model and obtain the sensory crispness value of the cantaloupe sample to be tested based on the multi-source information data; S4. Construct a sensory aroma dimension evaluation model and obtain the sensory aroma value of the cantaloupe sample to be tested based on the multi-source information data; S5. Construct a three-dimensional quality evaluation model and obtain the comprehensive quality score of the cantaloupe sample based on the sensory sweetness value, the sensory crispness value, and the sensory aroma value.

2. The method for three-dimensional quality evaluation of Hami melon based on multi-source information fusion according to claim 1, characterized in that: The multi-source information data includes the glucose content, sucrose content, fructose content, flesh firmness, flesh crispness, flesh compactness, and content of key aroma active ingredients of the cantaloupe sample to be tested.

3. The method for three-dimensional quality evaluation of Hami melon based on multi-source information fusion according to claim 2, characterized in that, The sensory sweetness dimension evaluation model includes a sensory sweetness value conversion formula: In the formula: The sensory sweetness value. For glucose content, This is the weighting coefficient for the sweetness of glucose. For sucrose content, This is the sweetness weighting coefficient for sucrose. For fructose content, This is the sweetness weighting coefficient for fructose, and The value is 0.

75. The value is 1.

0. The value is 1.

7.

4. The method for three-dimensional quality evaluation of Hami melon based on multi-source information fusion according to claim 2, characterized in that, The sensory crispness dimension evaluation model includes a sensory crispness value conversion formula: In the formula: Sensory crispness value, The firmness of the fruit flesh, This is a weighting coefficient for the crispness of the fruit flesh. The flesh is crisp. This is the weighting coefficient for the crispness of the fruit flesh. For the firmness of the fruit flesh, The crispness weighting coefficient represents the firmness of the fruit flesh. For the offset constant term, and The value is -0.00114. The value is 0.0008146. The value is -0.00146. The value is 2.7948.

5. The method for three-dimensional quality evaluation of Hami melon based on multi-source information fusion according to claim 2, characterized in that, The sensory aroma dimension evaluation model includes a sensory aroma value conversion formula: In the formula: The sensory aroma value, The first among the key aroma active ingredients The content of each component The first among the key aroma active ingredients Aroma weight coefficients corresponding to each component.

6. The method for three-dimensional quality evaluation of Hami melon based on multi-source information fusion according to claim 5, characterized in that, The key aroma active ingredients are obtained through the following steps: collecting volatile components of cantaloupe aroma compounds, calculating the odor activity value of each volatile component, and screening out volatile components with odor activity values ​​greater than or equal to 1 to obtain the key aroma active ingredients; wherein, the odor activity value is calculated as follows: In the formula: For the first The odor activity value of the volatile components, For the first The content of volatile components, For the first Odor threshold of a certain volatile component in water.

7. The three-dimensional quality evaluation method for Hami melon based on multi-source information fusion according to claim 6, characterized in that: The key aroma active ingredients include (Z)-2-nonenal, ethyl acetate, trans-2-,cis-6-nonadienal, methyl 2-methylbutyrate, 1-octen-3-ol, and ethyl 2-methylbutyrate.

8. The three-dimensional quality evaluation method for Hami melon based on multi-source information fusion according to claim 5, characterized in that, The aroma weight coefficients corresponding to the key aroma active ingredients are calculated and obtained through the following steps: A hierarchical structure is constructed based on analytic hierarchy process (AHP), comprising a target layer with the evaluation objective of superior cantaloupe aroma quality and a constraint layer with key aroma active ingredients as evaluation criteria; pairwise importance comparisons are performed on the key aroma active ingredients using scaling methods to construct a judgment matrix; a consistency check is performed on the obtained judgment matrix to ensure that it meets the consistency check criteria; weight column vectors are calculated through the judgment matrix to finally obtain the aroma weight coefficients corresponding to each key aroma active ingredient.

9. The three-dimensional quality evaluation method for Hami melon based on multi-source information fusion according to claim 1, characterized in that, The three-dimensional quality evaluation model includes a comprehensive quality score conversion formula: In the formula: To determine the overall quality score, The sensory sweetness value. The weighting coefficients corresponding to the sensory sweetness values. Sensory crispness value, The weighting coefficient corresponding to the sensory crispness value. The sensory aroma value, The weighting coefficients corresponding to the sensory aroma values. and 。 10. The method for three-dimensional quality evaluation of Hami melon based on multi-source information fusion according to claim 1, characterized in that, It also includes the following steps: S6. A grading standard is constructed based on the comprehensive quality score of the cantaloupe samples. The grading standard includes special grade, first grade and second grade.

Images