Food processing detection system based on big data

Abstract

The invention discloses a food processing detection system based on big data. The food processing detection system comprises a data acquisition layer, a data transmission layer and a server, the data acquisition layer comprises a sensor module, an image acquisition module and a manual input terminal; the data transmission layer is used for transmitting the data acquired by the data acquisition layer to a server; the server comprises a data storage module used for receiving data transmitted by the data transmission layer; the data classification module is used for classifying the data in the data storage module into non-alarm information data and alarm information data; relates to the technical field of food processing detection. A working process of an alarm analysis module can quickly determine abnormal related data causing an alarm. Through correlation analysis of historical alarm and non-alarm information data, abnormal related data can be quickly locked from related data, the troubleshooting time is greatly shortened, and losses and risks caused by food quality problems are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of food processing detection, and in particular to a food processing detection system based on big data. Background Art

[0002] In the food processing industry, as consumers' requirements for food safety and quality are increasing, ensuring strict monitoring and quality inspection of the food processing process has become a vital task. Traditional food processing inspection often relies on manual sampling inspection and simple instrument monitoring, which has many limitations.

[0003] Traditional food quality inspection methods, such as video surveillance and manual sampling, lack the ability to effectively utilize and analyze historical data when facing food quality issues. Past alarm information and non-alarm information data have not been systematically integrated and deeply mined, and cannot provide valuable reference for current quality inspection and problem diagnosis. This means that when companies face similar quality problems, they often need to repeat the same troubleshooting steps, and the troubleshooting efficiency needs to be improved. Summary of the invention

[0004] In order to solve the technical problems existing in the background technology, the present invention proposes a food processing detection system based on big data.

[0005] The present invention proposes a food processing detection system based on big data, comprising: a data collection layer, a data transmission layer and a server; The data acquisition layer includes: sensor module, image acquisition module and manual input terminal; Data transmission layer: used to transmit the data collected by the data collection layer to the server; The server includes: Data storage module: used to receive data transmitted by the data transmission layer; Data classification module: used to classify the data in the data storage module into: non-alarm information data and alarm information data; The non-alarm information data is further classified into: raw material data, processing technology data, and production environment data; Food detection alarm module: used to alarm when food detection does not meet the standards; Alarm analysis module: used to find out the data information that caused the alarm by analyzing the historical non-alarm information data and historical alarm information data in the data classification module.

[0006] Preferably, Sensor module: used to collect physical, chemical and biological parameter data during the processing, including temperature sensor, humidity sensor, pressure sensor, pH sensor, and microbial sensor; Image acquisition module: used to cover the food processing area with cameras, capture images of food raw materials, semi-finished products and finished products, and obtain data information on appearance characteristics, color and shape; Manual input terminal: A specially designed data entry interface is used for staff to manually input data that cannot be automatically collected by sensors and cameras.

[0007] Preferably, in the data classification module, Raw material data is further classified into: purchase information data, quality inspection report data, storage condition data, and other data; The processing technology data is further classified into: process information data, process parameter data, equipment information data, and other data; The production environment data is further classified into: workshop location data, environmental parameter data, and other data.

[0008] Preferably, in the alarm analysis module, the abnormality-related data related to the alarm information is found, and the process is as follows: Step 1: For the non-alarm information data and alarm information data in the data classification module, for each alarm information data, record the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data, and migrate the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data from the non-alarm information data to the alarm information data; Step 2: Next, data correlation analysis is performed, using statistical analysis methods to analyze the correlation between the raw material data, processing technology data, production environment data and the alarm information data corresponding to each alarm information data, and obtain the relevant data corresponding to several alarm information data; Step 3: Analyze the non-alarm information data to determine the range of normal data; Step 4: Secondly, the relevant data corresponding to the alarm information is brought into the range value of normal data for comparison, and abnormal relevant data that deviates from the range value of normal data is found. The obtained abnormal relevant data is the data information that causes the alarm; For the latest alarm information data provided by the food detection alarm module, the above steps 1, 2, 3 and 4 are used to obtain the data information causing the latest alarm.

[0009] Preferably, the server further comprises: Related information mining module: Using Apriori algorithm or FP-Growth algorithm, the historical data of the data storage module is analyzed, the association relationship between the data is mined, and the mined associated data information is stored in the associated information library.

[0010] Preferably, the server further comprises: Alarm display module: used to display the abnormality-related data provided by the alarm analysis module when the food detection alarm module alarms, and synchronously display the associated data of the alarm data abnormality-related data provided by the associated information mining module.

[0011] Preferably, the server further comprises: a correlation data analysis module: for the correlation data of the alarm data abnormality-related data provided by the correlation information mining module, data analysis is performed from the non-alarm information data to determine the range value of normal data, the correlation data is compared with the range value of normal data, and the correlation data deviating from the range value of normal data is found, if there is correlation data deviating from the range value of normal data, a correlation data report is generated and displayed together with the alarm display module; Preferably, in the alarm analysis module and the associated data analysis module, data analysis is performed from the non-alarm information data to determine the range of normal data as follows: using the quantile method, the 25% quantile Q1 and the 75% quantile Q3 are calculated, and the normal data range is determined to be: Q1±1.5×(Q3-Q1); If the data is a discrete variable, the quantile method is not used to determine the normal data range. At this time, it is necessary to count the frequency of discrete variables in non-alarm information data, calculate the median Z and interquartile range IQR, and determine the normal data range as: Z±1.5×IQR.

[0012] Preferably, a food processing detection method based on big data comprises the following steps: S1. Collect physical, chemical and biological parameter data during the processing process through the sensor module; use the image acquisition module to cover the food processing area with the camera, and take images of food raw materials, semi-finished products and finished products to obtain appearance characteristics, color and shape data information; Through the manual input terminal, the user manually inputs the data that cannot be automatically collected by sensors and cameras through the data entry interface; S2, using the data transmission layer to transmit the data collected by the data collection layer to the data storage module of the server for data storage; S3, the data classification module classifies the data in the data storage module into non-alarm information data and alarm information data; The non-alarm information data is further subdivided into: raw material data, processing technology data and production environment data; Raw material data is divided into purchase information data, quality inspection report data, storage condition data, and other data; Processing technology data is divided into process information data, process parameter data, equipment information data, and other data; Production environment data is divided into workshop location data, environmental parameter data, and other data; S4. When the food test does not meet the standards, the food test alarm module will sound an alarm; The alarm analysis module processes the alarm information: For the non-alarm information data and alarm information data in the data classification module, for each alarm information data, record the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data, and migrate the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data from the non-alarm information data to the alarm information data; Next, data correlation analysis is performed, using statistical analysis methods to analyze the correlation between the raw material data, processing technology data, production environment data and alarm information data corresponding to each alarm information data, and obtain the relevant data corresponding to several alarm information data; Perform data analysis on non-alarm information data to determine the range of normal data; Secondly, the relevant data corresponding to the alarm information is brought into the range value of normal data for comparison, and abnormal relevant data that deviates from the range value of normal data is found. The obtained abnormal relevant data is the data information that causes the alarm; For the latest alarm information data provided by the food detection alarm module, the step in S4 is used to obtain the data information causing the latest alarm; S5. Analyze the historical data of the data storage module using the Apriori algorithm or the FP-Growth algorithm, mine the association relationship between the data, and store the mined association data information in the association information library; For the associated data of the abnormal data related to the alarm data, the normal data range value is determined from the non-alarm information data, and the associated data is compared with the normal data range value. If there is associated data that deviates from the normal data range value, a related data report is generated.

[0013] S6. When the food detection alarm module alarms, the abnormality-related data provided by the alarm analysis module is displayed, and the associated data of the alarm data abnormality-related data provided by the associated information mining module and the associated data report generated by the associated data analysis module are simultaneously displayed.

[0014] In the present invention, the proposed food processing detection system based on big data has the following beneficial technical effects: 1. By analyzing the correlation between the raw material data, processing technology data, production environment data and alarm information data corresponding to each alarm information data, the relevant data corresponding to several alarm information data are obtained, and the relevant data are extracted from the numerous raw material data, processing technology data and production environment data to reduce the subsequent calculation amount, so that the workflow of the alarm analysis module can quickly determine the abnormal related data that causes the alarm.

[0015] 2. Through the correlation analysis of historical alarm and non-alarm information data, combined with data correlation analysis and comparison with normal data range values, abnormal related data can be quickly locked from related data, greatly shortening the troubleshooting time and reducing the losses and risks caused by food quality problems.

[0016] 3. Using Apriori algorithm or FP-Growth algorithm to mine related information can discover hidden associations between data. In the related data analysis module, if there is related data that deviates from the normal data range, a related data report is generated and displayed together with the alarm display module, which can provide food processing companies with more comprehensive problem diagnosis information so that more effective improvement strategies can be formulated later.

[0017] 4. The alarm display module will simultaneously display abnormal related data, associated data and associated data reports when an alarm is triggered, so that relevant personnel of food processing enterprises can obtain comprehensive alarm information at a glance. This helps enterprises to quickly organize response measures. Relevant departments can quickly coordinate raw material procurement, processing technology adjustments, and production environment improvement work based on the displayed data information, improve the response speed and coordination efficiency of enterprises in dealing with food quality issues, minimize the output rate of unqualified food, and protect consumer rights and corporate reputation.

[0018] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a principle block diagram of the system of the present invention; Figure 2 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0020] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar symbols throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0021] like Figure 1 A food processing detection system based on big data is shown, comprising: Data collection layer, data transmission layer and server; The data collection layer includes: Sensor module: used to collect physical, chemical and biological parameter data during the processing, including temperature sensor, humidity sensor, pressure sensor, pH sensor, and microbial sensor; Image acquisition module: used to cover the food processing area with cameras, capture images of food raw materials, semi-finished products and finished products, and obtain data information on appearance characteristics, color and shape for subsequent appearance inspection, foreign body recognition and other analyses; Food processing areas include raw material storage areas, processing production lines, and finished product packaging areas; Manual input terminal: Through a specially designed data input interface, staff can manually input data that cannot be automatically collected by sensors and cameras, such as raw material source information, processing parameters, additive usage, operator information, etc. Data transmission layer: used to transmit the data collected by the data collection layer to the server; The server includes: Data storage module: used to receive data transmitted by the data transmission layer; Data classification module: used to classify the data in the data storage module into: non-alarm information data and alarm information data; in, The alarm information data and non-alarm information data are further classified into: raw material data, processing technology data, production environment data In an optional embodiment, in the data classification module: Raw material data is further classified into: purchase information data, quality inspection report data, storage condition data, and other data; The processing technology data is further classified into: process information data, process parameter data, equipment information data, and other data; The production environment data is further classified into: workshop location data, environmental parameter data, and other data; Food detection alarm module: used to alarm when food detection does not meet the standards; How to detect whether food meets the standards can be done directly using existing technologies; Alarm analysis module: used to analyze alarm information data and find out abnormal data related to the alarm information. The abnormal data obtained is the data information that caused the alarm: In an optional embodiment, in the alarm analysis module, abnormality-related data related to the alarm information is found, and the process is as follows: Step 1: For the non-alarm information data and alarm information data in the data classification module, for each alarm information data, record the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data, and migrate the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data from the non-alarm information data to the alarm information data; Step 2: Next, data correlation analysis is performed, using statistical analysis methods to analyze the correlation between the raw material data, processing technology data, production environment data and the alarm information data corresponding to each alarm information data, and obtain the relevant data corresponding to several alarm information data; as follows: For continuous data, the Pearson correlation coefficient is calculated. When the Pearson correlation coefficient meets the set standard, it is judged as relevant data; When the data is a categorical variable, a contingency table is constructed, and then the chi-square test is used to determine whether there is an association between the variables. If the chi-square value of the chi-square test meets the set standard, it is judged to be related data.

[0022] Here, the relevant data is only the data associated with the alarm information data, but it does not mean that the relevant data is the cause of the alarm. The relevant data only means that the alarm information data may be related to it. Whether it is it or not requires further judgment.

[0023] The significance of relevant data judgment lies in the fact that there are a lot of data in raw material data, processing technology data, and production environment data. Relevant data can be extracted from these data to reduce the subsequent calculation amount.

[0024] Step 3: Analyze the non-alarm information data to determine the range of normal data; The range values ​​for normal data are determined as follows: The quantile method was used to calculate the 25% quantile Q1 and the 75% quantile Q3, and the normal data range was determined to be: Q1± 1.5×(Q3 - Q1); It should be noted that Q1 and Q3 are quartiles, Q2 is the 50% quantile, and Q2 is not used here.

[0025] If the data is a discrete variable, the quantile method is not used to determine the normal data range. At this time, it is necessary to count the frequency of discrete variables in non-alarm information data, calculate the median Z and interquartile range IQR, and determine the normal data range as: Z±1.5×IQR.

[0026] Step 4: Secondly, the relevant data corresponding to the alarm information is brought into the range value of normal data for comparison, and abnormal relevant data that deviates from the range value of normal data is found. The obtained abnormal relevant data is the data information that causes the alarm; For the latest alarm information data provided by the food detection alarm module, the above steps 1, 2, 3 and 4 are used to obtain the data information causing the latest alarm.

[0027] By analyzing the correlation between the raw material data, processing technology data, production environment data and the alarm information data corresponding to each alarm information data, the relevant data corresponding to several alarm information data are obtained. The relevant data are extracted from the numerous raw material data, processing technology data and production environment data to reduce the subsequent calculation amount, so that the workflow of the alarm analysis module can quickly determine the abnormal related data that caused the alarm.

[0028] Through the correlation analysis of historical alarm and non-alarm information data, combined with data correlation analysis and comparison with normal data range values, abnormal related data can be quickly identified from the relevant data, greatly shortening the troubleshooting time and reducing the losses and risks caused by food quality problems.

[0029] Association information mining module: Use Apriori algorithm or FP-Growth algorithm to analyze the historical data of the data storage module, mine the association between data, such as the association between raw material characteristics and processing parameters, the association between quality problems and production environment factors, etc., and store the mined association data information in the association information library; Apriori algorithm is a representative algorithm for association rule mining; FP-growth algorithm refers to Frequent Pattern-growth, which uses a compact data structure to store all the information needed to find frequent itemsets.

[0030] Using the Apriori algorithm or FP-Growth algorithm to mine related information can discover hidden associations between data. In the related data analysis module, if there is related data that deviates from the normal data range, a related data report is generated and displayed together with the alarm display module, which can provide food processing companies with more comprehensive problem diagnosis information so that more effective improvement strategies can be formulated later.

[0031] Alarm display module: used to display the abnormality-related data provided by the alarm analysis module when the food detection alarm module alarms, and synchronously display the associated data of the alarm data abnormality-related data provided by the associated information mining module.

[0032] The alarm display module will simultaneously display abnormal related data, associated data and associated data reports when an alarm is triggered, so that relevant personnel of food processing enterprises can obtain comprehensive alarm information at a glance. This helps enterprises to quickly organize response measures. Relevant departments can quickly coordinate raw material procurement, processing technology adjustments, and production environment improvement work based on the displayed data information, improve the response speed and coordination efficiency of enterprises in dealing with food quality issues, minimize the output rate of unqualified food, and protect consumer rights and corporate reputation.

[0033] Related data analysis module: For the related data of the alarm data abnormality related data provided by the related information mining module, data analysis is performed from the non-alarm information data to determine the range value of normal data, and the range value of the related data is compared with the range value of normal data to find out the related data that deviates from the range value of normal data. If there is related data that deviates from the range value of normal data, a related data report is generated and displayed together with the alarm display module.

[0034] Using the Apriori algorithm or FP-Growth algorithm to mine related information can discover hidden associations between data. In the related data analysis module, if there is related data that deviates from the normal data range, a related data report is generated and displayed together with the alarm display module, which can provide food processing companies with more comprehensive problem diagnosis information so that more effective improvement strategies can be formulated later.

[0035] like Figure 2 A food processing detection method based on big data is shown, comprising the following steps: S1. Collect physical, chemical and biological parameter data during the processing process through the sensor module; use the image acquisition module to cover the food processing area with the camera, and take images of food raw materials, semi-finished products and finished products to obtain appearance characteristics, color and shape data information; Through the manual input terminal, the user manually inputs the data that cannot be automatically collected by sensors and cameras through the data entry interface; S2, using the data transmission layer to transmit the data collected by the data collection layer to the data storage module of the server for data storage; S3, the data classification module classifies the data in the data storage module into non-alarm information data and alarm information data; The non-alarm information data is further subdivided into: raw material data, processing technology data and production environment data; Raw material data is divided into purchase information data, quality inspection report data, storage condition data, and other data; Processing technology data is divided into process information data, process parameter data, equipment information data, and other data; Production environment data is divided into workshop location data, environmental parameter data, and other data; S4. When the food test does not meet the standards, the food test alarm module will sound an alarm; The alarm analysis module processes the alarm information: For the non-alarm information data and alarm information data in the data classification module, for each alarm information data, record the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data, and migrate the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data from the non-alarm information data to the alarm information data; Next, data correlation analysis is performed, using statistical analysis methods to analyze the correlation between the raw material data, processing technology data, production environment data and alarm information data corresponding to each alarm information data, and obtain relevant data corresponding to several alarm information data; Perform data analysis on non-alarm information data to determine the range of normal data; Secondly, the relevant data corresponding to the alarm information is brought into the range value of normal data for comparison, and abnormal relevant data that deviates from the range value of normal data is found. The obtained abnormal relevant data is the data information that causes the alarm; For the latest alarm information data provided by the food detection alarm module, the step in S4 is used to obtain the data information causing the latest alarm; In an optional embodiment, the method for obtaining relevant data corresponding to a plurality of alarm information data is as follows: For continuous data, the Pearson correlation coefficient is calculated. When the Pearson correlation coefficient meets the set standard, it is judged as relevant data; When the data is a categorical variable, a contingency table is constructed, and then the chi-square test is used to determine whether there is an association between the variables. If the chi-square value of the chi-square test meets the set standard, it is judged to be related data.

[0036] Here, the relevant data is only the data associated with the alarm information data, but it does not mean that the relevant data is the cause of the alarm. The relevant data only means that the alarm information data may be related to it. Whether it is it or not requires further judgment.

[0037] The significance of relevant data judgment lies in the fact that there are a lot of data in raw material data, processing technology data, and production environment data. Relevant data can be extracted from these data to reduce the subsequent calculation amount.

[0038] In an optional embodiment, the range value of normal data is determined as follows: The quantile method was used to calculate the 25% quantile Q1 and the 75% quantile Q3, and the normal data range was determined to be: Q1± 1.5×(Q3 - Q1); It should be noted that Q1 and Q3 are quartiles, Q2 is the 50% quantile, and Q2 is not used here.

[0039] If the data is a discrete variable, the quantile method is not used to determine the normal data range. In this case, it is necessary to count the frequency of discrete variables in non-alarm information data, calculate the median Z and interquartile range IQR, and determine the normal data range as: Z±1.5×IQR; S5. Analyze the historical data of the data storage module using the Apriori algorithm or the FP-Growth algorithm, mine the association relationship between the data, and store the mined association data information in the association information library; For the associated data of the abnormal data related to the alarm data, the normal data range value is determined from the non-alarm information data, and the associated data is compared with the normal data range value. If there is associated data that deviates from the normal data range value, a related data report is generated.

[0040] S6. When the food detection alarm module alarms, the abnormality-related data provided by the alarm analysis module is displayed, and the associated data of the alarm data abnormality-related data provided by the associated information mining module and the associated data report generated by the associated data analysis module are simultaneously displayed.

[0041] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0042] In the embodiments provided by the present invention, it should be understood that the disclosed system or method can be implemented in other ways. For example, the above-described embodiments of the invention are only illustrative, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation.

[0043] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0044] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0045] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the basic features of the present invention.

[0046] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A food processing detection system based on big data, characterized in that: include: Data collection layer, data transmission layer and server; The data acquisition layer includes: sensor module, image acquisition module and manual input terminal; Data transmission layer: used to transmit the data collected by the data collection layer to the server; The server includes: Data storage module: used to receive data transmitted by the data transmission layer; Data classification module: used to classify the data in the data storage module into: non-alarm information data and alarm information data; The non-alarm information data is further classified into: raw material data, processing technology data, and production environment data; Food detection alarm module: used to alarm when food detection does not meet the standards; Alarm analysis module: used to find out the data information that caused the alarm by analyzing the historical non-alarm information data and historical alarm information data in the data classification module.

2. The food processing detection system based on big data according to claim 1 is characterized in that: In the data collection layer: Sensor module: used to collect physical, chemical and biological parameter data during the processing, including temperature sensor, humidity sensor, pressure sensor, pH sensor, and microbial sensor; Image acquisition module: used to cover the food processing area with cameras, capture images of food raw materials, semi-finished products and finished products, and obtain data information on appearance characteristics, color and shape; Manual input terminal: A specially designed data entry interface is used for staff to manually input data that cannot be automatically collected by sensors and cameras.

3. The food processing detection system based on big data according to claim 1, characterized in that: In the alarm analysis module, find out the abnormal data related to the alarm information. The process is as follows: Step 1: For the non-alarm information data and alarm information data in the data classification module, for each alarm information data, record the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data, and migrate the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data from the non-alarm information data to the alarm information data; Step 2: Next, data correlation analysis is performed, using statistical analysis methods to analyze the correlation between the raw material data, processing technology data, production environment data and the alarm information data corresponding to each alarm information data, and obtain the relevant data corresponding to several alarm information data; Step 3: Analyze the non-alarm information data to determine the range of normal data; Step 4: Secondly, the relevant data corresponding to the alarm information is brought into the range value of normal data for comparison, and abnormal relevant data that deviates from the range value of normal data is found. The obtained abnormal relevant data is the data information that causes the alarm; For the latest alarm information data provided by the food detection alarm module, the above steps 1, 2, 3 and 4 are used to obtain the data information causing the latest alarm.

4. The food processing detection system based on big data according to claim 3 is characterized in that: The server also includes: Related information mining module: Using Apriori algorithm or FP-Growth algorithm, the historical data of the data storage module is analyzed, the association relationship between the data is mined, and the mined associated data information is stored in the associated information library.

5. The food processing detection system based on big data according to claim 4 is characterized in that: The server also includes: Alarm display module: used to display the abnormality-related data provided by the alarm analysis module when the food detection alarm module alarms, and synchronously display the associated data of the alarm data abnormality-related data provided by the associated information mining module.

6. The food processing detection system based on big data according to claim 5 is characterized in that: The server also includes: Related data analysis module: For the related data of the alarm data abnormality related data provided by the related information mining module, data analysis is performed from the non-alarm information data to determine the range value of normal data, and the range value of the related data is compared with the range value of normal data to find out the related data that deviates from the range value of normal data. If there is related data that deviates from the range value of normal data, a related data report is generated and displayed together with the alarm display module.

7. The food processing detection system based on big data according to claim 6 is characterized in that: In the alarm analysis module and the associated data analysis module, data analysis is performed on the non-alarm information data to determine the range of normal data as follows: the quantile method is used to calculate the 25% quantile Q1 and the 75% quantile Q3, and the normal data range is determined to be: Q1±1.5×(Q3 - Q1); If the data is a discrete variable, the quantile method is not used to determine the normal data range. At this time, it is necessary to count the frequency of discrete variables in non-alarm information data, calculate the median Z and interquartile range IQR, and determine the normal data range as: Z±1.5×IQR.

8. The food processing detection method based on big data according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1, collect physical, chemical and biological parameter data during the processing through sensor modules; Use the image acquisition module to cover the food processing area with cameras to capture images of food raw materials, semi-finished products and finished products, thereby obtaining appearance characteristics, color and shape data information; Through the manual input terminal, the user manually inputs the data that cannot be automatically collected by sensors and cameras through the data entry interface; S2, using the data transmission layer to transmit the data collected by the data collection layer to the data storage module of the server for data storage; S3, the data classification module classifies the data in the data storage module into non-alarm information data and alarm information data; The non-alarm information data is further subdivided into: raw material data, processing technology data and production environment data; S4. When the food test does not meet the standards, the food test alarm module will sound an alarm; The alarm analysis module processes the alarm information: For the non-alarm information data and alarm information data in the data classification module, for each alarm information data, record the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data, and migrate the raw material data, processing technology data and production environment data in the non-alarm information data corresponding to each alarm information data from the non-alarm information data to the alarm information data; Next, data correlation analysis is performed, using statistical analysis methods to analyze the correlation between the raw material data, processing technology data, production environment data and alarm information data corresponding to each alarm information data, and obtain relevant data corresponding to several alarm information data; Perform data analysis on non-alarm information data to determine the range of normal data; Secondly, the relevant data corresponding to the alarm information is brought into the range value of normal data for comparison, and abnormal relevant data that deviates from the range value of normal data is found. The obtained abnormal relevant data is the data information that causes the alarm; For the latest alarm information data provided by the food detection alarm module, the step in S4 is used to obtain the data information causing the latest alarm; S5. Analyze the historical data of the data storage module using the Apriori algorithm or the FP-Growth algorithm, mine the association relationship between the data, and store the mined association data information in the association information library; For the associated data of the abnormal data related to the alarm data, determine the normal data range value from the non-alarm information data, compare the associated data with the normal data range value, and generate an associated data report if there is associated data that deviates from the normal data range value; S6. When the food detection alarm module alarms, the abnormality-related data provided by the alarm analysis module is displayed, and the associated data of the alarm data abnormality-related data provided by the associated information mining module and the associated data report generated by the associated data analysis module are simultaneously displayed.

Images