A method, system and device for food quality detection based on Monascus fermentation

Abstract

The present application discloses a method, system and equipment for food quality detection based on Monascus fermentation, which relates to the field of microbial technology. The method comprises: performing pre-treatment operations of washing, crushing and homogenizing food samples collected based on an efficiency model to obtain target detection substances; performing detection of pollutants and active ingredients on the target detection substances to generate detection data; and performing integrated analysis on the detection data to obtain quality detection results. Through the scheme of the present application, the detection quality can be improved and detailed analysis results can be provided.

Description

Technical Field

[0001] The present application relates to the field of microbial technology, and in particular to a method, system and equipment for food quality detection based on Monascus fermentation. Background Art

[0002] Monascus is a fungus used to make traditional fermented foods such as red yeast rice. Red yeast rice is widely used for its natural pigments and possible health benefits. Quality testing of foods fermented with Monascus is important because these products may contain secondary metabolites that cause health problems, such as aflatoxin, as well as substances that are beneficial to health, such as monascusin.

[0003] Current detection technologies face problems such as high cost, complex operation, and long detection cycle. These defects limit the rapid and large-scale detection of Monascus fermentation products, especially in resource-limited environments. To address these problems, some studies are exploring faster and more cost-effective methods. For example, near-infrared spectroscopy (NIR) technology and electronic nose technology have been proposed for rapid detection of food safety and quality. Although these methods provide the possibility of rapid detection, they may have problems with insufficient accuracy and inability to provide detailed chemical composition analysis.

[0004] Therefore, there is an urgent need for a technical solution that can improve the detection quality and provide detailed analysis results. Summary of the invention

[0005] In order to solve the deficiencies of the prior art, the embodiments of the present application provide a method, system and device for food quality detection based on Monascus fermentation. The present application solves the technical problems of the prior art such as low detection quality and single analysis results.

[0006] The embodiment of the present application provides a method for food quality detection based on Monascus fermentation, comprising: performing pre-treatment operations of washing, crushing and homogenizing a food sample collected based on an efficiency model to obtain a target detection substance; wherein the efficiency model includes: Where E represents sampling efficiency, n represents the number of samples collected, V represents the volume (or mass) of each sample, Q represents the representativeness and homogeneity of the number of samples and the volume or mass of each sample, and c represents the p represents fixed cost and variable cost, t p represents the processing time, and λ represents the weight coefficient for balancing the relationship between time cost and economic cost; the target detection substance is tested for pollutants and active ingredients to generate detection data; and the detection data is integrated and analyzed to obtain the quality detection result.

[0007] In a possible implementation, the detection of contaminants and active ingredients of the target detection substance to generate detection data includes: using enzyme-linked immunosorbent assay to detect specific microbial metabolites or contaminants; using Raman spectrometer for rapid screening to identify the type of Monascus and the structure of active substances in the sample; analyzing the content of each active ingredient by high performance liquid chromatography to obtain accurate quantitative values; using optical tweezers combined with Raman spectroscopy technology to identify Monascus strains to obtain food strain data; and splicing the detection results of enzyme-linked immunosorbent assay, Raman spectrometer, high performance liquid chromatography and optical tweezers combined with Raman spectroscopy technology to obtain detection data.

[0008] In a possible implementation, the integrated analysis of the detection data to obtain the quality detection result includes: performing data cleaning on the detection data to obtain a first feature; normalizing the first feature to obtain a second feature; calculating the statistical value of the second feature to obtain a quality stability result; combining the historical second feature with the current second feature, analyzing the trend in the generation process to obtain a quality pattern result; using a first machine learning algorithm to perform anomaly detection on the second feature to obtain a quality anomaly value result; using a second machine learning algorithm to predict the second feature to obtain a future quality trend result; and integrating and analyzing the quality stability result, the quality pattern result, the quality anomaly value result and the future quality trend result to obtain a quality detection result.

[0009] In a possible implementation, calculating the statistical value of the second feature to obtain a quality stability result includes: calculating the mean, median, mode, standard deviation and variance of the second feature to obtain a target statistic; and performing a hypothesis test based on the target statistic, wherein the hypothesis test is used to determine whether the differences between different batches are statistically significant to obtain a quality stability result.

[0010] In a possible implementation, the historical second feature is combined with the current second feature to analyze the trend in the generation process to obtain a quality model result, including: using a simple moving average or an exponential smoothing moving average to smooth the time series data of the second feature to obtain a first trend; applying a seasonal decomposition time series analysis of the seasonal pattern in the data to obtain a second trend; and splicing the first trend with the second trend to obtain a quality model result.

[0011] In a possible implementation, using the first machine learning algorithm to perform anomaly detection on the second feature to obtain a quality anomaly value result includes: based on historical data, using a random forest to mark the outliers of the second feature as the quality anomaly value result.

[0012] In one possible implementation, using a second machine learning algorithm to predict the second feature to obtain future quality trend results includes: using an autoregressive integrated moving average model or a seasonal ARIMA model to predict future values ​​to obtain a first prediction value; training a neural network model based on historical data, wherein the neural network model is used to predict future values ​​to obtain a second prediction value; and performing nonlinear addition of the first prediction value and the second prediction value to obtain future quality trend results.

[0013] In a possible implementation, the quality stability result, the quality pattern result, the quality anomaly result and the future quality trend result are integrated and analyzed to obtain the quality detection result, including: filling the quality stability result, the quality pattern result, the quality anomaly result and the future quality trend result into a preset prompt word template to obtain a target prompt word; and inputting the target prompt word into a large language model to obtain an analysis report of the quality detection result.

[0014] An embodiment of the present application also provides a system for food quality detection based on Monascus fermentation, comprising: a pre-treatment unit, a detection unit and an analysis unit; wherein the pre-treatment unit is used to perform pre-treatment operations of cleaning, crushing and homogenizing food samples collected based on an efficiency model to obtain target detection substances; the detection unit is used to detect pollutants and active ingredients in the target detection substances to generate detection data; and the analysis unit is used to perform integrated analysis of the detection data to obtain quality detection results.

[0015] An embodiment of the present application also provides a device for food quality detection based on Monascus fermentation, comprising: a processor, a memory, and a system bus; wherein the processor and the memory are connected via the system bus; the memory is used to store one or more programs, and the one or more programs include instructions, which, when executed by the processor, enable the processor to execute any one of the methods described in the above methods.

[0016] Through the method, system and equipment for food quality detection based on Monascus fermentation provided above, the embodiments of the present application can improve the detection quality and provide detailed analysis results by introducing multi-component joint analysis and language large model. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A schematic flow chart of a method for food quality detection based on Monascus fermentation provided in an embodiment of the present application;

[0019] Figure 2 A functional schematic diagram of a system for food quality detection based on Monascus fermentation provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless otherwise specifically stated.

[0021] Those skilled in the art can understand that the terms "first", "second" and the like in the embodiments of the present disclosure are only used to distinguish different steps, devices or modules, etc., and neither represent any specific technical meaning nor represent the necessary logical order between them. It should also be understood that in the embodiments of the present disclosure, "multiple" can refer to two or more, and "at least one" can refer to one, two or more. It should also be understood that for any component, data or structure mentioned in the embodiments of the present disclosure, in the absence of explicit limitation or contrary revelation given in the context, it can generally be understood as one or more. In addition, the term "and / or" in the present disclosure is only a description of the association relationship of the associated objects, indicating that there can be three relationships, for example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present disclosure generally indicates that the associated objects before and after are an "or" relationship. It should also be understood that the description of each embodiment in the present disclosure emphasizes the differences between the embodiments, and the same or similar parts can refer to each other. For the sake of brevity, they will not be repeated one by one.

[0022] At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present disclosure and its application or use. The techniques, methods and devices known to ordinary technicians in the relevant fields may not be discussed in detail, but where appropriate, the techniques, methods and devices should be considered part of the specification. It should be noted that similar numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0024] Figure 1 A schematic flow chart of a method for food quality detection based on Monascus fermentation provided in an embodiment of the present application.

[0025] like Figure 1 As shown, at step S101, the food sample collected based on the efficiency model is cleaned, crushed and homogenized to obtain the target detection substance. The collection of food samples is the starting point of the entire detection process. The sampling process must not only ensure the representativeness and uniformity of the sample, but also take into account cost-effectiveness. The sampling must be able to reflect the average quality of the entire production batch or processing batch. The sampling method based on the efficiency model can effectively balance the quality representativeness, cost and processing time of the sample. This model is particularly suitable for large-scale production environments, where the number and volume (or quality) of samples may vary from production batch to production batch. Collecting food samples includes collecting based on the efficiency model, and the efficiency model includes:

[0026]

[0027] Where E represents sampling efficiency, n represents the number of samples collected, V represents the volume (or mass) of each sample, Q represents the representativeness and homogeneity of the number of samples and the volume or mass of each sample, and c represents the p represents fixed cost and variable cost, t p represents the processing time, and λ represents the weight coefficient used to balance the relationship between time cost and economic cost.

[0028] After collection, the samples need to undergo pre-processing operations, including cleaning, crushing and homogenization, to obtain substances for target detection. The cleaning step removes surface impurities and possible contaminants in the sample, providing a clean basis for crushing and homogenization. Crushing is the process of breaking solid samples into smaller particles for more uniform mixing and more efficient chemical or biological analysis. During the homogenization process, the sample is further processed to make its components evenly distributed at the microscopic level, ensuring that each part of the sample can represent the average properties of the entire sample. Through the above steps, it can be ensured that the test substances obtained from the food samples are highly consistent and representative in quality and composition.

[0029] At step S102, the target detection material is subjected to the detection of pollutants and active ingredients to generate detection data. In this step, the target detection material obtained from the previous step is subjected to the detection of pollutants. Pollutants may include but are not limited to bacteria, fungi, heavy metals, residual pesticides, etc. The presence of these pollutants may have a negative impact on product quality and even pose a threat to consumer health. Available technology includes gas chromatography-mass spectrometry (GC-MS), liquid chromatography-mass spectrometry (LC-MS), atomic absorption spectroscopy (AAS) and high performance liquid chromatography (HPLC), etc. These methods can accurately identify and quantitatively analyze various pollutants to ensure the accuracy and reliability of the test results.

[0030] The detection of active ingredients is to confirm the presence and concentration of beneficial ingredients in Monascus fermented products, including monascusin, tannins, polyphenols and other substances with health functions. All test data will be collected and analyzed, including the type and content of contaminants and detailed information on active ingredients. These data not only provide direct evidence of product quality, but can also be used to screen strains that can produce high-quality and low-contamination products. By comparing the contaminant and active ingredient data of fermented products produced by different strains, the performance of each strain can be evaluated. Those strains that can maximize the yield of active ingredients and minimize the generation of contaminants are selected for further production and research.

[0031] Specifically, in one embodiment, it includes but is not limited to: using enzyme-linked immunosorbent assay to detect specific microbial metabolites or pollutants. Its operating principle is based on the specific binding of antibodies to antigens. This method can highly specifically identify and quantitatively analyze target compounds. The color change produced by the enzyme-labeled secondary antibody can intuitively read the content of pollutants or metabolites in the sample, thereby evaluating the sample quality.

[0032] A Raman spectrometer is used for rapid screening to identify the type of Monascus and the structure of the active substance in the sample. Raman spectroscopy is a non-destructive testing technology based on the change of the vibration energy level of the sample molecule, which can quickly identify the type of Monascus and the structure of its active substance in the sample. The advantage of this technology is its non-invasive and rapid response characteristics, which is suitable for rapid on-site detection. By analyzing the Raman spectrum, detailed information about the structure of chemical components can be obtained, which is of great value for distinguishing different Monascus strains and their metabolites.

[0033] The content of each active ingredient is analyzed by high performance liquid chromatography to obtain accurate quantitative values. High performance liquid chromatography is used to analyze the content of each active ingredient in the sample and provide accurate quantitative values. This technology can effectively separate and measure multiple components in complex samples through different chromatographic columns and detectors. The application of HPLC ensures the accuracy and repeatability of the test results, which is crucial to ensure product consistency and quality control.

[0034] Identify red yeast rice strains using optical tweezers combined with Raman spectroscopy to obtain food strain data. Optical tweezers is a technology that uses laser beams to capture tiny particles. When combined with Raman spectroscopy, it can perform fine spectral analysis at the individual strain level. This technology allows detection directly on living cells to obtain more biological information about the characteristics of the strain. By locating specific strains with optical tweezers and combining Raman spectroscopy analysis, high-quality red yeast rice strains can be identified very accurately, thereby obtaining relevant data on the strains on food.

[0035] and splicing the test results of enzyme-linked immunosorbent assay, Raman spectrometer, high performance liquid chromatography, and optical tweezers combined with Raman spectroscopy to obtain the test data.

[0036] At step S103, the test data is integrated and analyzed to obtain the quality test results. First, the test data is cleaned to obtain the first feature. Then the first feature is normalized to obtain the second feature. Data cleaning is the preliminary stage of data analysis, and its purpose is to remove noise and outliers in the test data to ensure data quality. In this process, those illogical data points (such as negative concentration values) will be eliminated, incorrect data inputs (such as unit errors) will be corrected, and missing values ​​will be filled. After obtaining the first feature, the next step is to normalize these features. The purpose of normalization is to eliminate the influence of data units and dimensions so that the data can be comparable under the same standard. Commonly used normalization methods include minimum-maximum standardization and Z score standardization (standard deviation standardization). Through normalization, the first feature can be converted into the second feature, providing a standardized data basis for subsequent statistical analysis.

[0037] After obtaining the second feature, the statistical value of the second feature is calculated to obtain the quality stability result. Including: calculating the mean, median, mode, standard deviation and variance of the second feature to obtain the target statistic. These statistics can reveal the central trend, degree of variation and distribution pattern of product quality. And based on the target statistic, a hypothesis test is performed, and the hypothesis test is used to determine whether the differences between different batches are statistically significant to obtain the quality stability result. Based on the above target statistics, a hypothesis test is further performed. This step is used to evaluate whether the product quality differences between different batches are statistically significant. Hypothesis testing methods include t-test, analysis of variance, etc. Through hypothesis testing, it can be determined whether each batch of products is statistically significantly different, thereby judging the quality stability of the product.

[0038] Next, the trend in the generation process is analyzed by combining the historical second feature with the current second feature to obtain a quality model result. In one embodiment, the method includes: smoothing the time series data of the second feature using a simple moving average or an exponential smoothing moving average to obtain a first trend; applying a seasonal decomposition time series to analyze the seasonal pattern in the data to obtain a second trend; and splicing the first trend with the second trend to obtain a quality model result.

[0039] First, the time series data of the second feature is smoothed, aiming to reduce random fluctuations and noise in the data so that the underlying trends of the data can be more clearly identified and analyzed. Smoothing can work in two ways: by calculating the average of a continuous fixed subset of data points, or by using an exponentially smoothed moving average to give more recent data a higher weight. Subsequently, seasonal decomposition techniques are applied to analyze seasonal patterns in the data. Seasonal decomposition is a method for analyzing time series data to identify and separate seasonal factors that affect the data. Seasonally Decomposed Time Series (STL) is a flexible seasonal decomposition method that allows time series to be seasonally adjusted while considering both trend and seasonal components. This method helps analysts understand the cyclical changes in data over different time periods by extracting seasonal components from the data. Finally, the first trend obtained by simple moving average or exponential smoothing is integrated with the second trend obtained by seasonal decomposition. Integrating these two trends can create a more comprehensive quality pattern result, which helps predict future quality trends and formulate corresponding quality control strategies.

[0040] Secondly, the second feature is detected for anomalies using the first machine learning algorithm to obtain a quality anomaly result, including: based on historical data, the outliers of the second feature are marked using a random forest as the quality anomaly result. Random forest is a machine learning method based on decision tree integration, which works by building multiple decision trees and outputting category patterns (classification problems) or average prediction values ​​(regression problems). In the present invention, random forest is used to identify and mark outliers in the second feature. The second feature is predicted using a second machine learning algorithm to obtain future quality trend results, including: predicting future values ​​using an autoregressive integral moving average model or a seasonal ARI MA model to obtain a first prediction value. A neural network model is trained based on historical data, and the neural network model is used to predict future values ​​to obtain a second prediction value. The first prediction value is nonlinearly added to the second prediction value to obtain a future quality trend result.

[0041] Finally, the quality stability result, the quality mode result, the quality anomaly value result and the future quality trend result are integrated and analyzed to obtain the quality detection result. In one embodiment, it includes: filling the quality stability result, the quality mode result, the quality anomaly value result and the future quality trend result into a preset prompt word template to obtain a target prompt word. And inputting the target prompt word into a large language model to obtain an analysis report of the quality detection result.

[0042] Specifically, descriptive target prompt words are generated by filling the integrated data into the preset prompt word template. These prompt words accurately outline the key information of each quality assessment result, providing a basis for subsequent in-depth analysis. The prompt word template is designed to automatically extract key information from the data and convert it into an easy-to-understand text description. Subsequently, the generated target prompt words are input into a large language model. These large pre-trained language models, such as GPT or BERT, can understand and process natural language and provide high-quality text generation services. The large language model will generate a detailed quality inspection result analysis report based on the input target prompt words. The report not only includes an assessment of the current product quality, but may also provide improvement suggestions, risk warnings, and predictions of future trends. Such an analysis report can help decision makers understand the overall situation of product quality and formulate more effective quality control strategies and production adjustments.

[0043] Figure 2A functional schematic diagram of a system for food quality detection based on Monascus fermentation provided in an embodiment of the present application. It should be understood that the system shown in the figure is exemplary and not restrictive. This means that the system architecture involved is not limited to a specific form or design, but is presented as an example. In other words, the architecture shown in the figure can be regarded as a way of expression to clearly describe related concepts and relationships, and does not exclude other forms of architecture. Therefore, when interpreting the architecture in the picture, it should be understood that the model is flexible and diverse, and its purpose is to provide an exemplary description rather than a restrictive provision on a specific form.

[0044] Specifically, the embodiment of the present application includes: a pre-processing unit, a detection unit and an analysis unit; wherein the pre-processing unit 201 is used to perform pre-processing operations such as cleaning, crushing and homogenization on the food samples collected based on the efficiency model to obtain the target detection substance; the detection unit 202 is used to detect pollutants and active ingredients in the target detection substance to generate detection data; and the analysis unit 203 is used to perform integrated analysis on the detection data to obtain the quality detection result.

[0045] Furthermore, an embodiment of the present application also provides a device for food quality detection based on Monascus fermentation, comprising: a processor, a memory, and a system bus; the processor and the memory are connected via the system bus; the memory is used to store one or more programs, and the one or more programs include instructions, which, when executed by the processor, enable the processor to execute any of the above methods.

[0046] Furthermore, an embodiment of the present application also provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute any one of the detection methods of the above-mentioned processing methods.

[0047] It can be known from the description of the above implementation mode that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., including several instructions for enabling a computer device (which can be a personal computer, a server, or a network communication device such as a media gateway, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.

[0048] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0049] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0050] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for food quality detection based on Monascus fermentation, characterized in that: include: The food samples collected based on the efficiency model are cleaned, crushed and homogenized to obtain the target detection substance; wherein the efficiency model includes: ， in, represents the sampling efficiency, represents the number of samples collected, represents the volume of each sample, Indicates the representativeness and homogeneity of the sample quantity and volume or mass of each sample. represents fixed costs and variable costs, Indicates the processing time, Represents the weight coefficient used to balance the relationship between time cost and economic cost; Testing the target test substance for contaminants and active ingredients to generate test data; and Conduct integrated analysis on the test data to obtain quality test results; The detection of pollutants and active ingredients of the target test substance to generate test data includes: Detection of specific microbial metabolites or contaminants using enzyme-linked immunosorbent assays; Rapid screening using Raman spectroscopy to identify the species of Monascus and the structure of active substances in the sample; The content of each active ingredient is analyzed by high performance liquid chromatography to obtain accurate quantitative values; Using optical tweezers combined with Raman spectroscopy to identify red yeast rice strains to obtain food strain data; and splicing the test results of enzyme-linked immunosorbent assay, Raman spectrometer, high-performance liquid chromatography and optical tweezers combined with Raman spectroscopy to obtain test data; Perform integrated analysis on the test data to obtain quality test results, including: Performing data cleaning on the detection data to obtain a first feature; Normalizing the first feature to obtain the second feature; Calculating the statistical value of the second characteristic to obtain a quality stability result; Combine the historical second characteristics with the current second characteristics to analyze the trend in the generation process to obtain quality model results; performing anomaly detection on the second feature using the first machine learning algorithm to obtain a quality outlier result; A second machine learning algorithm is used to predict the second feature to obtain a future quality trend result; and the quality stability result, the quality pattern result, the quality outlier result and the future quality trend result are integrated and analyzed to obtain a quality detection result.

2. The method according to claim 1, characterized in that: Calculating the statistical value of the second feature to obtain the quality stability result includes: The mean, median, mode, standard deviation and variance of the second feature are calculated to obtain a target statistic; and a hypothesis test is performed based on the target statistic, wherein the hypothesis test is used to determine whether the difference between different batches is statistically significant to obtain a quality stability result.

3. The method according to claim 1, characterized in that The historical second feature is combined with the current second feature to analyze the trend in the generation process to obtain the quality model results including: Use simple moving average or exponential moving average to smooth the time series data of the second feature to obtain the first trend; Apply seasonal decomposition to analyze seasonal patterns in time series data to obtain secondary trends; and splice the primary trend with the secondary trend to obtain a quality pattern result.

4. The method according to claim 1, characterized in that The first machine learning algorithm is used to perform anomaly detection on the second feature to obtain a quality anomaly value result including: Based on historical data, outliers of the second feature are marked using random forest as quality outlier results.

5. The method according to claim 1, characterized in that The second feature is predicted using the second machine learning algorithm to obtain future quality trend results including: Forecast future values ​​using an autoregressive integrated moving average model or a seasonal ARIMA model to obtain a first forecast value; A neural network model is trained based on historical data, the neural network model is used to predict future values ​​to obtain a second predicted value; and the first predicted value and the second predicted value are nonlinearly added to obtain a future quality trend result.

6. The method according to claim 1, characterized in that The quality stability result, the quality mode result, the quality abnormal value result and the future quality trend result are integrated and analyzed to obtain the quality detection result including: Fill the quality stability result, the quality mode result, the quality anomaly result and the future quality trend result into a preset prompt word template to obtain a target prompt word; and input the target prompt word into a large language model to obtain an analysis report of the quality detection result.

7. A food quality detection system based on Monascus fermentation, used to implement the method according to any one of claims 1 to 6, characterized in that: include: Pre-processing unit, detection unit and analysis unit; wherein, The pre-processing unit is used to perform pre-processing operations of cleaning, crushing and homogenizing the food samples collected based on the efficiency model to obtain the target detection substance; The detection unit is used to detect pollutants and active ingredients in the target detection substance to generate detection data; And the analysis unit is used to perform integrated analysis on the detection data to obtain the quality detection result.

8. A food quality detection device based on Monascus fermentation, characterized in that: include: A processor, a memory, and a system bus; wherein the processor and the memory are connected via the system bus; The memory is used to store one or more programs, wherein the one or more programs include instructions, and when the instructions are executed by the processor, the processor executes the method according to any one of claims 1 to 6.

Images