Method for rapidly identifying different processing modes and host plant cynomorium songaricum based on electronic tongue technology

Abstract

The invention relates to the technical field of cynomorium songaricum identification, in particular to a method for rapidly identifying cynomorium songaricum of different processing modes and host plants based on an electronic tongue technology. The method comprises the following steps: weighing 1.0 g of a sample, adding 10mL of water, dissolving, filtering after 30min, repeating for three times, merging filtrate, obtaining a test solution, and detecting by using an electronic tongue; statistical analysis; the method has the beneficial effects or advantages of rapid and accurate quality identification, non-destructive detection, diversified processing modes and host plant identification, promotion of healthy development of the cynomorium songaricum industry, improvement of intelligent and automatic levels and the like. Due to the advantages, the method has wide application prospects and market potential in the field of cynomorium songaricum quality identification.

Description

Technical Field

[0001] The invention relates to the technical field of Cynomorium songaricum identification, and specifically to a method for rapidly identifying Cynomorium songaricum with different processing methods and host plants based on electronic tongue technology. Background Art

[0002] Taste refers to the feeling caused by soluble substances entering the mouth. It is a comprehensive reflection of the cooperation and coordination of various flavor substances and is an important component of flavor. Electronic tongue technology has gradually emerged in the past 10 years. As an innovative taste quality detection technology, it is based on a multi-sensor array and can obtain the taste characteristics of liquid samples and analyze the taste differences of samples. It has a similar perception threshold and taste intensity recognition ability as humans, thereby ensuring the accuracy and authenticity of the test results, making it have significant advantages in taste quality detection and can be used in many fields such as food, medicine and agriculture. Therefore, electronic tongue can achieve efficient and reliable rapid detection of taste quality. This experiment used electronic tongue technology to detect and analyze Cynomorium songaricum from different host plants and processed in different ways, in order to provide a solid theoretical basis for the rapid quality identification and characteristic flavor evaluation of Cynomorium songaricum based on taste.

[0003] At present, the electronic tongue has been widely used in the detection of food flavor quality, but it has not yet been applied to Cynomorium songaricum, and there is no application for related invention patents.

[0004] In summary, the existing technology has the following problems: traditional quality identification methods are often time-consuming, highly subjective, and difficult to ensure the accuracy of the results. There is a lack of fast and effective methods and technical means for Cynomorium songaricum quality identification. The introduction of electronic tongue technology is expected to solve this problem and promote the accuracy and efficiency of Cynomorium songaricum quality identification. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a method for quickly identifying different processing methods and host plants Cynomorium songaricum based on electronic tongue technology. The method can quickly and accurately identify different processing methods and host plants Cynomorium songaricum, and can quickly and objectively evaluate the quality of Cynomorium songaricum, providing a scientific decision-making basis for each link in the industrial chain.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for quickly identifying different processing methods and host plants of Cynomorium songaricum based on electronic tongue technology, characterized in that it comprises the following steps:

[0008] Step 1, sample processing; the different host plants of the samples are Cynomorium songaricum and Cynomorium serratum; the different processing methods of the samples include the following: (1) 3 slice thicknesses: 0.5 cm, 1 cm, and 2 cm thick fleshy stem slices; (2) 6 drying methods: oven 105°C for 30 min, drying at 65°C to constant weight; oven 105°C for 30 min, drying at 85°C to constant weight; oven 65°C for constant weight; oven 85°C for constant weight; vacuum freeze drying: 0.096hP, -100°C for constant weight; natural sun drying: placed outdoors for natural sun drying to constant weight; the 3 slice thicknesses and 6 drying methods are combined into 18 processing methods; the sample is crushed with a high-speed crusher and passed through an 80-mesh sieve for subsequent analysis;

[0009] Step 2: Weigh 1.0g of sample and add 10mL of water to dissolve it. Filter after 30 minutes. Repeat 3 times. Combine the filtrate to obtain the test solution and then use the electronic tongue to detect it. When testing, first use distilled water to clean the sensor for 120s, then test for 180s. Distilled water should be used to clean the sensor between different samples. The injection bottle specification is 10mL, and the test data is saved.

[0010] Step 3. Statistical analysis: The electronic tongue sensor response value data was used as the independent variable, and the processing method and host plant were used as the dependent variables. The data were shuffled, 80% of them were randomly selected as the training set, and the remaining 20% ​​were used as the test set. A Gaussian naive Bayes (GNB) model was established, and the machine learning algorithm was used to further accurately identify different categories of Cynomorium samples; each measurement indicator had 3 replicates, and the obtained data were analyzed and processed using SPSSPRO software and Origin 2022 software.

[0011] The present invention has the following beneficial effects:

[0012] 1. Rapid and accurate quality identification: The present invention uses electronic tongue technology to quickly and accurately identify the quality of Cynomorium songaricum in a short time. Compared with traditional manual identification or chemical analysis methods, this method greatly improves work efficiency, reduces human errors, and ensures the accuracy of identification results.

[0013] 2. Non-destructive testing: Using electronic tongue technology for quality identification, there is no need to perform destructive treatment on the Cynomorium songaricum samples, thus maintaining the integrity and original flavor of the samples. This non-destructive testing method not only reduces resource waste, but also provides more possibilities for subsequent research and product development.

[0014] 3. Diversified processing methods and host plant identification: The method of the present invention can effectively distinguish the effects of different processing methods and different host plants on the quality of Cynomorium songaricum. This is of great significance for consumers to choose high-quality Cynomorium songaricum products, and for manufacturers to optimize processing technology and select high-quality host plants.

[0015] 4. Promote the healthy development of the Cynomorium songaricum industry: Through rapid and accurate quality identification, the present invention helps to protect the rights and interests of consumers and improve consumers' trust in Cynomorium songaricum products. At the same time, it also provides strong technical support for the healthy development of the Cynomorium songaricum industry and promotes the development of the entire industry towards high quality and high added value.

[0016] 5. Improvement of intelligence and automation: The introduction of the present invention improves the intelligence and automation level of Cynomorium songaricum quality identification, reduces labor costs, and improves work efficiency. At the same time, it also provides intelligent data support for subsequent quality control and product development.

[0017] In summary, the beneficial effects or advantages of the present invention are mainly reflected in rapid and accurate quality identification, non-destructive testing, diversified processing methods and host plant identification, promotion of the healthy development of the Cynomorium songaricum industry, and improvement of the level of intelligence and automation. These advantages make the present invention have broad application prospects and market potential in the field of Cynomorium songaricum quality identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 The electronic tongue radar fingerprints of the odor characteristics of Cynomorium songaricum in 18 processing methods in the present invention;

[0019] Figure 2 The confusion matrix heat map of the training set (A) and test set (B) results of the GNB model for 18 processing methods in the present invention;

[0020] Figure 3 The confusion matrix heat map of the training set (A) and test set (B) results of the GNB model for three slice thicknesses in the present invention;

[0021] Figure 4 The confusion matrix heat map of the training set (A) and test set (B) results of the GNB model for the six drying methods in the present invention;

[0022] Figure 5 The confusion matrix heat map of the training set (A) and test set (B) results of the GNB model for the three types of drying methods in the present invention;

[0023] Figure 6 The electronic tongue radar fingerprints of the odor characteristics of different host plants Cynomorium songaricum in the present invention;

[0024] Figure 7This is a heat map of the confusion matrix of the training set (A) and test set (B) results of the GNB model for different host plants in the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Embodiment 1

[0027] 1. Sample processing

[0028] Different processing methods: (1) 3 slice thicknesses: 0.5 cm, 1 cm, and 2 cm thick fleshy stem slices. (2) 6 drying methods: oven 105°C for 30 min, drying at 65°C to constant weight; oven 105°C for 30 min, drying at 85°C to constant weight; oven 65°C to constant weight; oven 85°C to constant weight; vacuum freeze drying: 0.096hP, -100°C to constant weight; natural sun drying: placed outdoors to dry naturally to constant weight. Each treatment is numbered, where letters A to C represent slice thickness, numbers 1 to 6 represent drying methods, and the combination of numbers and letters represents processing methods. The specific numbers are shown in Table 1-1.

[0029] Different host plants: Take fresh Cynomorium songaricum and Cynomorium songaricum, cut the fleshy stems into 0.5 cm thick slices, and vacuum freeze-dry them.

[0030] The above samples were crushed with a high-speed crusher and passed through an 80-mesh sieve for subsequent analysis.

[0031] Table 1 Slice thickness, drying method and processing method number

[0032]

[0033]

[0034] 2 Instruments

[0035] Table 2 Test instruments

[0036]

[0037] 3 Experimental methods

[0038] Weigh 1.0g sample and add 10mL water to dissolve. Filter after 30min. Repeat 3 times. Combine the filtrate to obtain the test solution. When testing, first use distilled water to clean the sensor for 120s, then test for 180s. Distilled water should be used to clean the sensor between different samples. The injection bottle specification is 10mL, and the test data is saved.

[0039] 4 Statistical analysis

[0040] The electronic tongue sensor response value data was used as the independent variable, and the processing method and host plant as the dependent variables. The data were shuffled, and 80% of them were randomly selected as the training set and the remaining 20% ​​as the test set. A Gaussian naive Bayes (GNB) model was established, and the machine learning algorithm was used to further accurately identify different categories of Cynomorium songaricum samples.

[0041] Each measurement indicator was repeated three times, and the obtained data were analyzed and processed using SPSSPRO and Origin 2022 software.

[0042] 4.1 Identification of Cynomorium songaricum with different processing methods based on electronic tongue technology

[0043] 4.1.1 Analysis of electronic tongue response values

[0044] from Figure 1 It can be seen that the 27 electronic tongue sensors all respond to the flavor substances of Cynomorium songaricum processed in different ways, but there are differences in the response characteristics of different sensors, and the contribution to the test data is also different. The response value of C2 is generally greater than that of other treatments, and the response value of A4 is generally less than that of other treatments. There are 9 sensors that have the greatest impact on the response value of Cynomorium songaricum flavor, namely S4, S5, S10, S11, S13, S14, S15, S26, and S27 sensors. In summary, it shows that there are differences in the taste of the finished products of Cynomorium songaricum after being processed by different methods.

[0045] 4.1.2 Establishment of Gaussian Naive Bayes (GNB) classification model based on electronic tongue

[0046] 4.1.2.1 Establishment of GNB classification model based on 18 processing methods of electronic tongue

[0047] Depend on Figure 2 A shows that all training set samples are predicted correctly; Figure 2B shows that there are 9 samples in the test set that are correctly predicted, namely A1, A4, A6, B2, B3, B5, C2, C3 and C6; there are only 2 samples that are incorrectly predicted, namely B1 and C5. As shown in Table 4-1, the accuracy, precision and recall of this model on the training set are all 100.0%, and the F1-Score is 1.000, which performs very well. The accuracy, recall and precision obtained on the test set are all 81.8%, and the F1-Score is 0.818, and the evaluation effect on the test set is also good. Therefore, the 18 processing methods GNB discrimination model established based on the electronic tongue sensor data can accurately predict the Cynomorium samples with different processing methods.

[0048] Table 4-1 GNB model evaluation results of 18 processing methods

[0049]

[0050] (2) Establishment of GNB classification model for three slice thicknesses

[0051] Depend on Figure 3 A shows that there are 36 samples predicted correctly in the training set, including 11 0.5 cm slice samples, 11 1 cm slice samples, and 14 2 cm slice samples; there are 7 samples predicted incorrectly, including 4 0.5 cm slice samples and 3 1 cm slice samples. Figure 3 B shows that all samples in the test set are predicted correctly, with a total of 11 samples, including 3 0.5 cm slice samples, 4 1 cm slice samples and 4 2 cm slice samples. As shown in Table 4-2, the accuracy and recall rate of this model on the training set are both 83.7%, the precision rate is 84.1%, and the F1-Score is 0.627, which is a good performance. The accuracy, recall rate and precision rate obtained on the test set are all 100%, and the F1-Score is 1.000. The evaluation effect is very good, indicating that the three slice thickness GNB discrimination models established based on the electronic tongue sensor data can accurately predict Cynomorium samples with different slice thicknesses.

[0052] Table 4-2 Evaluation results of GNB model for three slice thicknesses

[0053]

[0054] (3) Establishment of GNB classification model for six drying methods based on electronic tongue

[0055] Depend on Figure 4A shows that there are 15 samples correctly predicted in the GNB model training set of 6 drying methods, including 5 samples of 105℃ fixation for 30min and 65℃ drying, 1 sample of 105℃ fixation for 30min and 85℃ drying, 3 samples of 65℃ drying, 2 samples of 85℃ drying and 4 freeze-dried samples; there are 28 samples wrongly predicted, including 1 sample of 105℃ fixation for 30min and 65℃ drying, 4 samples of 105℃ fixation for 30min, 5 samples of 65℃ drying, 6 samples of 85℃ drying, 9 samples of sun drying and 3 samples of freeze drying. Figure 4 B shows that there are 6 samples correctly predicted in the test set of the 6 drying methods GNB model, including 1 sample of 105℃ fixation for 30min and 65℃ drying, 2 samples of 105℃ fixation for 30min and 85℃ drying, 1 sample of 85℃ drying and 2 freeze-dried samples; there are 5 samples wrongly predicted, including 2 samples of 105℃ fixation for 30min and 65℃ drying, 2 samples of 105℃ fixation for 30min and 85℃ drying and 1 sample of 65℃ drying. As shown in Table 4-3, the accuracy and recall rate of this model on the training set are 34.9%, the precision rate is 37.1%, and the F1-Score is 0.298, which is average. The accuracy and recall rate obtained on the test set are 54.5%, the precision rate is 68.2%, and the F1-Score is 0.564, which is a good evaluation effect, indicating that the 6 drying methods GNB model established based on the electronic tongue sensor data has a good performance in distinguishing the 6 drying methods on the test set.

[0056] Table 4-3 Evaluation results of GNB model for 6 drying methods

[0057]

[0058] (4) Establishment of a GNB classification model for three types of drying methods based on the electronic tongue

[0059] Depend on Figure 5 A shows that there are 21 samples correctly predicted in the GNB model training set of the three drying methods, including 14 drying samples, 3 sun-dried samples and 4 freeze-dried samples; there are 22 samples wrongly predicted, including 13 drying samples, 6 sun-dried samples and 3 freeze-dried samples. Figure 5B shows that there are 9 samples correctly predicted in the test set of the 6 drying methods GNB model, including 7 dried samples and 2 freeze-dried samples; there are 2 samples that are incorrectly predicted, both of which are dried samples. As shown in Table 4-4, the accuracy and recall rate of this model on the training set are both 48.8%, the precision rate is 61.6%, and the F1-Score is 0.523. The accuracy and recall rate on the test set are both 81.8%, the precision rate is 93.9%, and the F1-Score is 0.861, which is a good performance, indicating that the 3 types of drying methods GNB discrimination model established based on the electronic tongue sensor data can accurately predict the 3 types of samples with different drying methods.

[0060] Table 4-4 GNB model evaluation results for three types of drying methods

[0061]

[0062] 4.2 Identification of Cynomorium songaricum from different host plants based on electronic tongue technology

[0063] 4.2.1 Analysis of electronic tongue response values

[0064] from Figure 6 It can be seen that the 27 electronic tongue sensors have different response values ​​to the taste substances of different host plants Cynomorium. Overall, the electronic tongue response value of Cynomorium sargentii is greater than that of Cynomorium nitratum. Among the 27 sensors, 10 sensors S6, S7, S8, S9, S15, S16, S19, S20, S21, and S22 have a greater impact on the response value of Cynomorium odor, indicating that they play an important role in distinguishing the odor characteristics of different host plants Cynomorium. In summary, it shows that Cynomorium from different host plants has differences in odor.

[0065] 4.2.2 Establishment of Gaussian Naive Bayes (GNB) classification model based on electronic tongue

[0066] Depend on Figure 7 As can be seen from A, all samples in the GNB model training set of different host plants are predicted correctly, and Figure 7 B shows that all samples in the test set model are also predicted correctly. As shown in Table 4-5, the accuracy, precision and recall of this model on the training set and the test set are all 100.0%, and the F1-Score is 1.000, which is very good and the evaluation effect is very ideal. Therefore, the two host plant GNB discrimination models established based on the electronic tongue sensor data can accurately predict the Cynomorium samples of different host plants.

[0067] Table 4-5 Evaluation results of GNB model for different host plants

[0068]

[0069] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A method for quickly identifying different processing methods and host plants of Cynomorium songaricum based on electronic tongue technology, characterized in that: The steps include: Step 1: Sample processing; Step 2: Weigh 1.0g of sample and add 10mL of water to dissolve it. Filter after 30min. Repeat 3 times. Combine the filtrates to obtain the test solution and then use the electronic tongue to detect it. Step 3: Statistical analysis.

2. A method for rapidly identifying different processing methods and host plants of Cynomorium songaricum based on electronic tongue technology according to claim 1, characterized in that: In step one, the different host plants of the samples are Cynomorium songaricum and Cynomorium songaricum.

3. A method for rapidly identifying different processing methods and host plants of Cynomorium songaricum based on electronic tongue technology according to claim 1, characterized in that: In step 1, the different processing methods of the sample include the following: (1) 3 slice thicknesses: 0.5 cm, 1 cm, and 2 cm thick fleshy stem slices; (2) 6 drying methods: oven sterilization at 105°C for 30 min, drying at 65°C to constant weight; oven sterilization at 105°C for 30 min, drying at 85°C to constant weight; oven drying at 65°C to constant weight; oven drying at 85°C to constant weight; vacuum freeze drying: 0.096hP, -100°C to constant weight; natural sun drying: placed outdoors to dry naturally to constant weight; the 3 slice thicknesses and 6 drying methods are combined into 18 processing methods.

4. A method for rapidly identifying different processing methods and host plants of Cynomorium songaricum based on electronic tongue technology according to claim 1, characterized in that: In step 1, the sample was crushed with a high-speed crusher and passed through an 80-mesh sieve for subsequent analysis.

5. The method for quickly identifying different processing methods and host plants of Cynomorium songaricum based on electronic tongue technology according to claim 1, characterized in that: In step 2, the sensor is first cleaned with distilled water for 120 seconds and then tested for 180 seconds. The sensor needs to be cleaned with distilled water between testing different samples. The specification of the injection bottle is 10 mL, and the test data is saved.

6. A method for rapidly identifying different processing methods and host plants of Cynomorium songaricum based on electronic tongue technology according to claim 1, characterized in that: In step three, the electronic tongue sensor response value data were used as independent variables, and the processing method and host plant were used as dependent variables. The data were shuffled, and 80% of them were randomly selected as the training set and the remaining 20% ​​as the test set. A Gaussian naive Bayes (GNB) model was established to further accurately identify different categories of Cynomorium songaricum samples through machine learning algorithms.

7. A method for rapidly identifying different processing methods and host plants of Cynomorium songaricum based on electronic tongue technology according to claim 1, characterized in that: In step three, each measurement indicator was repeated three times, and the obtained data were analyzed and processed using SPSSPRO software and Origin 2022 software.

Images