Food conflict system based on AI screening
Through the AI-based food conflict system, using sensors and machine learning models to detect food risks in real time, the problem of high difficulty in food detection in the existing technology is solved, the efficiency and safety of food production are improved, and the production process is optimized.
Patent Information
- Application Number
- CN202411319953.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-23
- Publication Date
- 2024-12-10
AI Technical Summary
When existing food testing systems detect unqualified products, it is difficult to quickly identify the problem, resulting in an increase in time and labor costs, which increases the difficulty of error elimination.
Adopting AI-based food conflict system, including data acquisition module, model building module, risk assessment module, decision support module and feedback update module, provides real-time detection of nutritional components and potential safety risks of food through sensors, cameras and machine learning models, providing Production improvement suggestions and real-time early warnings to ensure food safety and production efficiency.
Real-time risk detection and problem prevention in food production have been achieved, the difficulty of eliminating problems has been reduced, production efficiency and food safety have been improved, production processes have been optimized, and waste and losses have been reduced.
Smart Images
Figure CN119107099A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of food production, and in particular to a food conflict system based on AI screening. Background Art
[0002] The food conflict system based on AI screening is a system that uses artificial intelligence technology to detect and prevent food conflicts. It is mainly used to identify potential conflicts in the food production process and provide corresponding solution suggestions.
[0003] However, the existing technology has the following shortcomings: Existing systems for testing food can only determine that a product is unqualified when it detects unqualified products. When investigating errors, they can only gradually eliminate the drawbacks on the production line. When the problem is detected, more time and manpower costs are consumed, which increases the difficulty of finding production errors. Summary of the invention
[0004] The purpose of the present invention is to provide a food conflict system based on AI screening to solve the problem that the existing system for detecting food can only judge that the product is unqualified when unqualified products are detected, and can only gradually eliminate the drawbacks on the production line when investigating the errors. When the problem is detected, more time and manpower costs are consumed, thereby increasing the difficulty of querying the production errors.
[0005] To achieve the above object, the present invention provides the following technical solutions: The present invention is a food conflict system based on AI screening, comprising: Data acquisition module, model building module, risk assessment module, decision support module and feedback update module; The risk assessment module includes a nutritional component analysis unit and a food conflict detection unit. The nutritional component analysis unit is used to analyze the nutritional components of food and generate a dietary assessment report to ensure that the food meets nutritional standards. The food conflict detection unit detects potential food safety risks, such as cross-contamination of allergens and improper ingredient combinations, based on a known food conflict knowledge base and machine learning models.
[0006] As a further solution of the present invention: the data acquisition module includes a data acquisition unit, a data preprocessing unit and a feature extraction unit; The data acquisition unit collects data at various stages of pre-prepared food production through sensors, cameras and other monitoring equipment, including raw material quality, production environment parameters (such as temperature and humidity), and operating steps, and extracts relevant data from online food databases and internal electronic records of the enterprise in order to fully understand the production process of pre-prepared foods.
[0007] As a further solution of the present invention: the data preprocessing unit is used to clean the collected data, remove noise and redundant information, and ensure the quality of the input data. The feature extraction unit is used to extract the chemical composition characteristics of the food, such as allergens and microbial content. These characteristics are particularly important for identifying food conflicts.
[0008] As a further solution of the present invention: the model building module includes building a learning model, model training and tuning, and integrated learning; Constructing a learning model is to select the most suitable machine learning model based on the characteristics of food ingredient data. Model training and tuning is to use the training data set to train the selected model and optimize the model performance by adjusting the hyperparameters. Ensemble learning combines the prediction results of multiple models to further improve the accuracy and robustness of the overall system.
[0009] As a further solution of the present invention: the decision support module includes providing production improvement suggestions, real-time warning and visual display; Providing production improvement suggestions is to provide specific handling suggestions for detected food conflicts, such as adjusting production parameters and changing raw material suppliers. Real-time warning is to establish a real-time warning system to immediately notify producers when potential food conflicts are detected, thereby reducing food safety risks.
[0010] As a further solution of the present invention: the visual display includes a data encryption unit, a privacy protection unit and a compliance checking unit; The data encryption unit is used to ensure the security of all user data during transmission and storage to prevent data leakage. The privacy protection unit adopts strict privacy protection policies and technical measures to protect users' personal information from being abused.
[0011] As a further aspect of the present invention, the compliance checking unit is used to ensure that the system complies with all relevant food safety regulations and standards, so as to provide legal and compliant recommendations.
[0012] As a further solution of the present invention: the feedback update module includes a cross-validation unit, a performance indicator unit and a continuous iteration unit; The cross-validation unit is used to evaluate the generalization ability of the model using the cross-validation method to avoid overfitting problems.
[0013] As a further solution of the present invention: the performance indicator unit uses multiple performance indicators such as accuracy, recall rate, and F1 score to comprehensively evaluate the effect of the model.
[0014] As a further solution of the present invention: the continuous iteration unit continuously adjusts and optimizes the model according to the results of the effect evaluation to achieve continuous iterative improvement of the model.
[0015] Compared with the prior art, the present invention has the following beneficial effects: Through the set risk assessment module, potential food safety risks such as allergen cross-contamination and improper ingredient combinations are detected based on the known food conflict knowledge base and machine learning model, so that errors in food production can be understood in real time and potential problems can be prevented. This not only improves production efficiency, but also reduces the difficulty of troubleshooting food production problems.
[0016] Based on the first beneficial effect, by setting up a decision support module, not only can the producer be notified immediately when a potential food conflict is detected to reduce food safety risks, but specific handling suggestions can also be provided for the detected food conflicts, such as replacing raw material suppliers or adjusting production steps. This not only improves the quality and safety of food production, but also optimizes the production process and reduces food waste and loss. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 It is a structural block diagram of the food conflict system based on AI screening of the present invention.
[0018] Figure 2 It is a structural block diagram of the data acquisition module of the present invention.
[0019] Figure 3 It is a schematic diagram of the specific steps of the food conflict system based on AI screening of the present invention. DETAILED DESCRIPTION
[0020] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.
[0021] See also Figures 1 to 3 ,In an embodiment of the present invention, a food conflict system based on AI screening includes: a data acquisition module, a model building module, a risk assessment module, a decision support module, and a feedback update module; The risk assessment module includes a nutrient analysis unit and a food conflict detection unit. The nutrient analysis unit is used to analyze the nutrient composition of food and generate a dietary assessment report to ensure that the food meets the nutritional standards. The food conflict detection unit detects potential food safety risks, such as allergen cross-contamination and improper ingredient combinations, based on a known food conflict knowledge base and a machine learning model. The risk assessment module is set up not only to ensure the nutritional value of the food produced, but also to detect the production process. When the detection is inconsistent with the prescribed standards, the abnormality can be discovered and handled in time, reducing food loss and waste caused by incorrect production.
[0022] The data acquisition module includes a data acquisition unit, a data preprocessing unit and a feature extraction unit; The data collection unit collects data at various stages of the production of pre-prepared dishes through sensors, cameras and other monitoring equipment, including raw material quality, production environment parameters (such as temperature and humidity), and operating steps, and extracts relevant data from online food databases and internal electronic records of the enterprise, so as to fully understand the production process of pre-prepared dishes; The data acquisition unit is set up to collect real-time data on the production environment and process, and the data extracted from the food database and the company's internal electronic records can provide a more comprehensive understanding of the production process and process requirements of pre-prepared dishes.
[0023] The data preprocessing unit is used to clean the collected data, remove noise and redundant information, and ensure the quality of input data. The feature extraction unit is used to extract the chemical composition characteristics of food, such as allergens and microbial content. These characteristics are particularly important for identifying food conflicts; The data preprocessing unit is used to more accurately ensure the quality of input data, and standardize the data to make it suitable for subsequent machine learning algorithms, thereby improving the accuracy and robustness of the model. The feature extraction unit not only analyzes the chemical composition of food, but also analyzes production parameters such as temperature and pressure, thereby reducing feature dimensions and retaining only the most influential features to improve the training efficiency and prediction accuracy of the model.
[0024] The model building module includes building learning models, model training and tuning, and integrated learning; Constructing a learning model is to select the most suitable machine learning model according to the characteristics of food ingredient data. Model training and tuning is to train the selected model using the training data set and optimize the model performance by adjusting the hyperparameters. Ensemble learning combines the prediction results of multiple models to further improve the accuracy and robustness of the overall system. First, select a suitable machine learning algorithm based on the type and scale of the data, and train the model using known pre-prepared food production data to allow the model to learn the rules and patterns in food production, so that it can identify potential conflicts and wrong steps.
[0025] The decision support module includes providing production improvement suggestions, real-time warnings and visual displays; Providing production improvement suggestions is to provide specific treatment suggestions for detected food conflicts, such as adjusting production parameters and changing raw material suppliers. Real-time warning is to establish a real-time warning system to immediately notify producers when potential food conflicts are detected, thereby reducing food safety risks. By providing production improvement suggestions and adjusting the accuracy of production parameters, when raw materials do not meet actual production standards, raw materials that meet production standards can be replaced in time to ensure food safety. The real-time early warning system can remind producers to assist producers in making decisions, and the visual display shows test results, risk assessment and decision support information through a graphical interface to improve user experience.
[0026] The visual display includes data encryption unit, privacy protection unit and compliance inspection unit; The data encryption unit is used to ensure the security of all user data during transmission and storage to prevent data leakage, while the privacy protection unit adopts strict privacy protection policies and technical measures to protect users' personal information from being abused; The data encryption unit is set up to encode the data to ensure that only users who have the key can access the data. This technology can effectively prevent data from being intercepted and read by unauthorized third parties during transmission, greatly enhancing the confidentiality and integrity of the data. The privacy protection unit can reduce the risk of illegal access and leakage of personal data and improve user satisfaction when using the system.
[0027] The compliance inspection unit is used to ensure that the system complies with all relevant food safety regulations and standards to provide legal and compliance recommendations; The compliance inspection unit can conduct compliance inspections regularly to promptly identify and correct potential compliance issues, thereby reducing risks in food production and ensuring that the food production process can operate continuously and stably.
[0028] The feedback update module includes a cross-validation unit, a performance indicator unit, and a continuous iteration unit; The cross-validation unit is used to evaluate the generalization ability of the model using the cross-validation method to avoid overfitting problems; The cross-validation unit divides the dataset into multiple subsets for training and validation. Cross-validation helps the model better adapt to unseen data, thereby improving the generalization ability of the model in practical applications, thereby improving the stability and reliability of the model.
[0029] The performance indicator unit uses multiple performance indicators such as accuracy, recall, and F1 score to comprehensively evaluate the effect of the model; The performance indicator unit can evaluate the quantitative method of model performance from different angles, making the model evaluation more comprehensive and accurate.
[0030] The continuous iteration unit continuously adjusts and optimizes the model based on the results of the effect evaluation to achieve continuous iterative improvement of the model; The continuous iteration unit achieves a gradual improvement in model performance by continuously adjusting model parameters and structure. During the iteration process, it can promote continuous improvement of models and algorithms to maintain the accuracy and relevance of the model.
[0031] Working principle: The specific process of system operation is as follows: S1. First, use cameras and sensors to collect food images and related data, such as food ingredients and production process parameters, and extract data from online food databases and electronic records to provide a rich source of information for subsequent analysis; remove noise and redundant information from the collected data, and extract features useful for food conflict detection to ensure the quality of input data; extract chemical composition features in food, such as allergens and microbial content; S2. Select appropriate machine learning algorithms based on the type and size of data, and train the models with known food production data to allow the models to learn the laws and patterns in food production, thereby identifying potential conflicts and wrong steps; S3. Detect potential food safety risks, such as allergen cross-contamination and improper ingredient combinations, based on a known food conflict knowledge base and machine learning models; S4. Provide specific treatment suggestions for detected food conflicts and establish a real-time early warning system. When abnormal situations occur, display the test results, risk assessment and decision support information through a graphical interface; S5. Regularly update the system based on new research results and user feedback to maintain system accuracy and practicality.
[0032] What is described above 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 conflict system based on AI screening, characterized in that: include: Data acquisition module, model building module, risk assessment module, decision support module and feedback update module; The risk assessment module includes a nutritional component analysis unit and a food conflict detection unit. The nutritional component analysis unit is used to analyze the nutritional components of food and generate a dietary assessment report to ensure that the food meets nutritional standards. The food conflict detection unit detects potential food safety risks, such as cross-contamination of allergens and improper ingredient combinations, based on a known food conflict knowledge base and machine learning models.
2. A food conflict system based on AI screening according to claim 1, characterized in that: The data acquisition module includes a data acquisition unit, a data preprocessing unit and a feature extraction unit; The data acquisition unit collects data at various stages of pre-prepared food production through sensors, cameras and other monitoring equipment, including raw material quality, production environment parameters (such as temperature and humidity), and operating steps, and extracts relevant data from online food databases and internal electronic records of the enterprise in order to fully understand the production process of pre-prepared foods.
3. The food conflict system based on AI screening according to claim 2, characterized in that: The data preprocessing unit is used to clean the collected data, remove noise and redundant information, and ensure the quality of the input data. The feature extraction unit is used to extract the chemical composition characteristics of food, such as allergens and microbial content. These characteristics are particularly important for identifying food conflicts.
4. The food conflict system based on AI screening according to claim 1, characterized in that: The model building module includes building a learning model, model training and tuning, and integrated learning; Constructing a learning model is to select the most suitable machine learning model based on the characteristics of food ingredient data. Model training and tuning is to use the training data set to train the selected model and optimize the model performance by adjusting the hyperparameters. Ensemble learning combines the prediction results of multiple models to further improve the accuracy and robustness of the overall system.
5. The food conflict system based on AI screening according to claim 1, characterized in that: The decision support module includes providing production improvement suggestions, real-time warning and visual display; Providing production improvement suggestions is to provide specific handling suggestions for detected food conflicts, such as adjusting production parameters and changing raw material suppliers. Real-time warning is to establish a real-time warning system to immediately notify producers when potential food conflicts are detected, thereby reducing food safety risks.
6. The food conflict system based on AI screening according to claim 5, characterized in that: The visual display includes a data encryption unit, a privacy protection unit and a compliance inspection unit; The data encryption unit is used to ensure the security of all user data during transmission and storage to prevent data leakage. The privacy protection unit adopts strict privacy protection policies and technical measures to protect users' personal information from being abused.
7. The food conflict system based on AI screening according to claim 6, characterized in that: The compliance checking unit is used to ensure that the system complies with all relevant food safety regulations and standards to provide legal and compliance recommendations.
8. The food conflict system based on AI screening according to claim 1, characterized in that: The feedback update module includes a cross-validation unit, a performance indicator unit and a continuous iteration unit; The cross-validation unit is used to evaluate the generalization ability of the model using the cross-validation method to avoid overfitting problems.
9. The food conflict system based on AI screening according to claim 8, characterized in that: The performance indicator unit uses multiple performance indicators such as accuracy, recall rate, and F1 score to comprehensively evaluate the effect of the model.
10. The food conflict system based on AI screening according to claim 8, characterized in that: The continuous iteration unit continuously adjusts and optimizes the model according to the results of the effect evaluation to achieve continuous iterative improvement of the model.