CN117153338A - Deep learning-based seasoning intake determining method and system - Google Patents
Deep learning-based seasoning intake determining method and system Download PDFInfo
- Publication number
- CN117153338A CN117153338A CN202310817345.5A CN202310817345A CN117153338A CN 117153338 A CN117153338 A CN 117153338A CN 202310817345 A CN202310817345 A CN 202310817345A CN 117153338 A CN117153338 A CN 117153338A
- Authority
- CN
- China
- Prior art keywords
- data
- user
- taken
- dishes
- flavoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000013135 deep learning Methods 0.000 title claims abstract description 21
- 235000011194 food seasoning agent Nutrition 0.000 title claims description 46
- 235000016709 nutrition Nutrition 0.000 claims abstract description 40
- 238000012549 training Methods 0.000 claims abstract description 25
- 230000035764 nutrition Effects 0.000 claims abstract description 20
- 235000013305 food Nutrition 0.000 claims description 20
- 235000019640 taste Nutrition 0.000 claims description 17
- 239000000126 substance Substances 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 14
- 201000010099 disease Diseases 0.000 claims description 12
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 12
- 239000004278 EU approved seasoning Substances 0.000 claims description 11
- 238000010411 cooking Methods 0.000 claims description 11
- 238000012544 monitoring process Methods 0.000 claims description 10
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 claims description 8
- 238000004519 manufacturing process Methods 0.000 claims description 8
- 238000002360 preparation method Methods 0.000 claims description 7
- 235000014347 soups Nutrition 0.000 claims description 7
- 208000004262 Food Hypersensitivity Diseases 0.000 claims description 5
- 206010016946 Food allergy Diseases 0.000 claims description 5
- 235000020932 food allergy Nutrition 0.000 claims description 5
- 206010067484 Adverse reaction Diseases 0.000 claims description 4
- 230000006838 adverse reaction Effects 0.000 claims description 4
- 229910021529 ammonia Inorganic materials 0.000 claims description 4
- -1 hydrogen ions Chemical class 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 235000000346 sugar Nutrition 0.000 claims description 4
- 230000036772 blood pressure Effects 0.000 claims description 3
- 239000000796 flavoring agent Substances 0.000 claims description 3
- 235000019634 flavors Nutrition 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 239000008280 blood Substances 0.000 claims description 2
- 210000004369 blood Anatomy 0.000 claims description 2
- 229910052739 hydrogen Inorganic materials 0.000 claims description 2
- 239000001257 hydrogen Substances 0.000 claims description 2
- 238000000354 decomposition reaction Methods 0.000 claims 2
- 235000021186 dishes Nutrition 0.000 description 26
- 230000002354 daily effect Effects 0.000 description 15
- 239000000463 material Substances 0.000 description 10
- 235000013311 vegetables Nutrition 0.000 description 8
- 235000013409 condiments Nutrition 0.000 description 7
- 230000036541 health Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 235000005911 diet Nutrition 0.000 description 5
- 206010020772 Hypertension Diseases 0.000 description 4
- 230000000378 dietary effect Effects 0.000 description 4
- 235000006694 eating habits Nutrition 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000013178 mathematical model Methods 0.000 description 4
- 239000002994 raw material Substances 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 239000011229 interlayer Substances 0.000 description 3
- 235000012054 meals Nutrition 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 235000002639 sodium chloride Nutrition 0.000 description 3
- 208000002720 Malnutrition Diseases 0.000 description 2
- 235000013527 bean curd Nutrition 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013124 brewing process Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000011217 control strategy Methods 0.000 description 2
- 230000001276 controlling effect Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 235000020805 dietary restrictions Nutrition 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 229940079593 drug Drugs 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 235000019197 fats Nutrition 0.000 description 2
- 235000013373 food additive Nutrition 0.000 description 2
- 239000002778 food additive Substances 0.000 description 2
- 235000000824 malnutrition Nutrition 0.000 description 2
- 230000001071 malnutrition Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 235000015097 nutrients Nutrition 0.000 description 2
- 208000015380 nutritional deficiency disease Diseases 0.000 description 2
- 150000003839 salts Chemical class 0.000 description 2
- 235000012046 side dish Nutrition 0.000 description 2
- 208000018556 stomach disease Diseases 0.000 description 2
- 241000894006 Bacteria Species 0.000 description 1
- 208000012895 Gastric disease Diseases 0.000 description 1
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 1
- 208000008589 Obesity Diseases 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 241000700605 Viruses Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 239000008186 active pharmaceutical agent Substances 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 208000022531 anorexia Diseases 0.000 description 1
- 235000014633 carbohydrates Nutrition 0.000 description 1
- 150000001720 carbohydrates Chemical class 0.000 description 1
- 230000001143 conditioned effect Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 206010061428 decreased appetite Diseases 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 206010012601 diabetes mellitus Diseases 0.000 description 1
- 230000037213 diet Effects 0.000 description 1
- 235000021196 dietary intervention Nutrition 0.000 description 1
- 235000008242 dietary patterns Nutrition 0.000 description 1
- 230000029087 digestion Effects 0.000 description 1
- 210000002249 digestive system Anatomy 0.000 description 1
- 238000004851 dishwashing Methods 0.000 description 1
- 235000005686 eating Nutrition 0.000 description 1
- 235000021191 food habits Nutrition 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 210000000987 immune system Anatomy 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 208000017169 kidney disease Diseases 0.000 description 1
- 238000009533 lab test Methods 0.000 description 1
- 239000010410 layer Substances 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 235000004213 low-fat Nutrition 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 235000003715 nutritional status Nutrition 0.000 description 1
- 235000020824 obesity Nutrition 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 239000003755 preservative agent Substances 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 235000019600 saltiness Nutrition 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 159000000000 sodium salts Chemical class 0.000 description 1
- 238000003756 stirring Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/60—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to nutrition control, e.g. diets
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Epidemiology (AREA)
- Computational Linguistics (AREA)
- Primary Health Care (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Nutrition Science (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention provides a flavoring intake determining method and system based on deep learning, which belongs to the field of deep learning and automatic fine control, and comprises the following steps: acquiring physical data of a user and nutritional information data ingested on the same day; training an LSTM model using the physical data of the user and the daily intake nutritional data; organizing physical data of a user and daily intake nutrition information into sequence data, taking one day as one time step, taking each time step as input of an LSTM model, and simultaneously using data within three days as the context of the input data; the LSTM model is used to output the individual flavoring amounts that the user needs to ingest the next day. The system is capable of controlling the daily intake of the flavoring of the user according to his physical and nutritional conditions.
Description
Technical Field
The invention belongs to the field of deep learning and automatic fine control, and particularly relates to a flavoring intake determining method and system based on deep learning.
Background
The elderly eat condiments with excessive amounts of high sodium salts, sugar, fat and carbohydrates, which may exacerbate basic diseases such as hypertension, diabetes, kidney disease, obesity and digestive system problems. Depending on the difference in region, culture, habit and education level, the proportions of men and women who retire the cooking skills among the elderly may be different. Generally, in the past social habits and cultural concepts, women are responsible for cooking and housekeeping matters in families in a higher proportion, while men are mainly engaged in outgoing work and economic income sources, so the old people tend to go out for dining after retirement.
In addition, problems that may occur with the home diet after retirement of the elderly include eating habits changes, malnutrition, food safety issues, kitchen safety issues, and social issues. For example, elderly people may prefer certain foods, lack necessary nutrients, and cause malnutrition due to taste preference, anorexia, etc. The immune system of the elderly is relatively weak, and bacteria and viruses in food are easy to infect; meanwhile, the elderly may have low tolerance to chemicals such as additives and preservatives in foods, and may easily cause food allergy and other food safety problems. In addition, after the elderly retired, life style and eating habits may be changed, for example, irregular eating patterns, too late eating time, etc., which easily affect digestion and absorption functions.
When the dish making robot in the prior art is used for making dishes, the fixed amount of seasoning can be added according to the dishes, and the amount of the seasoning can not be controlled according to the physical state of a user.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a flavoring intake determining method and system based on deep learning.
In order to achieve the above object, the present invention provides the following technical solutions:
a deep learning based seasoning intake determination method and system, comprising:
acquiring body data of a historical user and nutritional information data ingested on the same day;
predicting the total amount of each flavoring to be taken on the next day based on the body data of the historical user and the daily intake nutritional data; and taking the predicted total amount of each flavoring to be taken on the next day, the physical data of the historical user and the daily nutritional data as a training set;
training an LSTM model by using body data of a historical user in the training set and daily intake nutrition data, wherein when the total amount of each seasoning output by the LSTM model is the same as the predicted total amount of each seasoning to be taken in the next day, the LSTM model is trained;
the body data of the user and the daily intake nutrition information are formed into sequence data, one time step is taken as one day, each time step is taken as the input of the LSTM model, and the data in three days are taken as the context of the input data, so that the consumption of each flavoring needed to be taken by the user in the next day is output.
Further, the body data of the user includes: age, sex, height, weight, blood pressure, blood sugar, type of disease of the user, severity of disease of the user, known food allergy or adverse reaction, and medical treatment condition.
Further, the nutritional information that the user ingests daily is input into the matrix factorization model, and the food and dishes that are most suitable for the user are recommended by the matrix factorization model.
A deep learning based flavor intake determination system, comprising:
the data acquisition module is used for acquiring physical data of a user and nutritional information data ingested on the same day;
the main control module is used for:
predicting the total amount of each flavoring to be taken on the next day based on the body data of the historical user and the daily intake nutritional data; and taking the predicted total amount of each flavoring to be taken on the next day, the physical data of the historical user and the daily nutritional data as a training set;
training an LSTM model by using body data of a historical user in the training set and daily intake nutrition data, wherein when the total amount of each seasoning output by the LSTM model is the same as the predicted total amount of each seasoning to be taken in the next day, the LSTM model is trained;
the body data of the user and the daily intake nutrition information are formed into sequence data, one time step is taken as one day, each time step is taken as the input of the LSTM model, and the data in three days are taken as the context of the input data, so that the consumption of each flavoring needed to be taken by the user in the next day is output.
Further, the method further comprises the following steps:
the real-time monitoring module is used for monitoring chemical characteristic data of dishes in real time in the dish manufacturing process;
the main control module inputs the real-time chemical characteristic data of the dishes into a trained mlp neural network according to the real-time chemical characteristic data of the dishes, the mlp neural network model predicts the tastes of the dishes in real time, and the addition amount and the proportion of the seasonings are controlled according to the difference between the predicted result of the tastes of the dishes and the target taste value.
Further, the real-time monitoring module includes:
the pH sensor is used for detecting the concentration of hydrogen ions in dishes;
the salinity sensor is used for detecting the concentration of salt ions in dishes;
the ammonia sensor is used for monitoring the ammonia concentration in dishes.
Further, the method further comprises the following steps: a dish preparation module, comprising:
the inner container is used for cooking dishes or soup.
Further, the main control module includes: and the fuzzy controller and the neural network controller are connected in series to form a hybrid controller.
The invention provides a flavoring intake determining method and system based on deep learning, which comprises the following steps of
The beneficial effects are that:
the invention collects the physical data and nutrition intake information of the user, gives the total amount of the seasonings which the user can ingest every day according to the physical data and the nutrition intake information as a training set for training the LSTM neural network, and utilizes the LSTM neural network to output the total amount of the seasonings which the user can ingest the next day. The problem that in the prior art, a dish making robot can only add a fixed amount and proportion of seasonings according to dishes when making dishes and cannot control the amount of the seasonings according to the physical state of a user is solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention and the design thereof, the drawings required for the embodiments will be briefly described below. The drawings in the following description are only some of the embodiments of the present invention and other drawings may be made by those skilled in the art without the exercise of inventive faculty.
Fig. 1 is a schematic diagram of a deep learning-based seasoning method according to the present invention.
FIG. 2 is a schematic diagram of a seasoning zoned pretreatment and storage standby according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a dish making module according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a fresh food feeding and processing module according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a food material side dish and a stir-frying process in an embodiment of the invention.
Detailed Description
The present invention will be described in detail below with reference to the drawings and the embodiments, so that those skilled in the art can better understand the technical scheme of the present invention and can implement the same. The following examples are only for more clearly illustrating the technical aspects of the present invention, and are not intended to limit the scope of the present invention.
Examples:
the invention provides a flavoring intake determining method and system based on deep learning, particularly as shown in fig. 1, comprising the following steps: acquiring physical data of a user and nutritional information data ingested on the same day; giving the total amount of each flavoring to be taken on the next day by a doctor according to the physical data of the user and the daily nutritional data, and taking the total amount of each flavoring and the physical data of the user and the daily nutritional data as a training set; training an LSTM model by using body data of a user in the training set and daily intake nutrition data, wherein when the total amount of each flavoring outputted by the LSTM model is the same as the total amount of each flavoring which should be taken in the next day according to the body data of the user and the daily intake nutrition data by a doctor, the LSTM model training is completed; organizing physical data of a user and daily intake nutrition information into sequence data, taking one day as one time step, taking each time step as input of an LSTM model, and simultaneously using data within three days as the context of the input data; the LSTM model is used to output the individual flavoring amounts that the user needs to ingest the next day.
The following are details of the implementation of the system of the present invention:
1) Automated seasoning and manufacturing system:
the system consists of a seasoning box, an inner container, a sensor and a dish preparation system;
the seasoning box is designed according to the property of the seasoning, so that the seasoning box is convenient to store, and before the seasoning box is used, the seasoning can be added into an interlayer of the seasoning box, and the adding dosage is regulated and controlled by a control system; the inner container comprises a dish frying inner container and a soup cooking inner container, and can be used for cooking dishes and soup, and the size and the quantity of the inner containers are designed according to different price and requirements; the sensor is used for monitoring real-time information in the cooking or soup cooking process. And (5) automatically cleaning after finishing one dish each time, and transferring the finished dish to an interlayer for heat preservation. The dish preparation system is used for washing dishes and cutting vegetables.
After dish preparation system side dish is accomplished, move the real dish that holds edible material to the system execution layer by edible material delivery system like the height that the inner bag was located in the dish, add edible material in proper order according to the order such as dish and the order of adding the condiment. The subsequent stir-frying action is completed by the stir-frying liner.
2) A data storage and processing system:
basic information such as dish names, required food materials and quantity, manufacturing steps, cooking time and temperature, possible nutrition information, raw material sources, dish pictures, xiaoshi and the like of various menus is input and stored as a database 1.
The basic file of the physical health of the user is recorded, wherein the file key comprises dietary data such as the disease type, the severity of the disease, known food allergy or adverse reaction, drug treatment condition and the like of the user, and data items with risks and relativity in the physical examination data of the old are extracted and stored as a database 2.
And establishing a matching model by utilizing a machine learning algorithm according to the body information of the user. The user's physical condition and nutritional information is input into a machine learning algorithm to build a matching model to determine which foods and dishes are most suitable for the user. The model can be adjusted based on the health objectives, taste preferences, eating habits, disease types, etc. of the user; targeted dietary advice is provided based on the user's personal conditions (e.g., preferences and dietary restrictions), such as providing recipes, cooking advice, and health meal selection advice, among others.
The food record and the body data of the old are used as sequence data, the sequence data are input into an LSTM model, the total amount of the seasonings required to be ingested by a user in one day is output through training the LSTM model, and the seasonings required to be ingested by the user in one day are respectively placed in each interlayer of the seasoning box; in the process of making dishes by the dish making module, controlling the seasonings in the seasoning box to be added into the dishes according to a fixed proportion.
The process of the software terminal communicating with the user is embodied in the terminal interaction of the product, such as in the form of WeChat applet, public number, payment applet, etc.
3) And (3) a control system:
in a preferred embodiment, the inputs and outputs of the control system are defined, the inputs being defined as the proportions and properties of the various raw materials and condiments, and the outputs being properties of the conditioned condiments, such as taste, acidity, sweetness, saltiness, etc.
In a preferred scheme, a mathematical model of the control system is established according to defined input and output, a mass conservation equation set containing various raw materials, seasonings and mixtures is established through principles of mass conservation, mass balance and the like, and the mathematical model is established by using the equation set. And designing a control strategy of a control system according to the established mathematical model, using a fuzzy controller to realize fuzzy control of the system, using a neural network controller to realize intelligent control of the system, and connecting the fuzzy controller and the neural network controller in series to form a hybrid controller to realize intelligent and personalized control of the system.
In a preferred scheme, a single-chip microcomputer or a microprocessor is used for building the controller, and a programming language is used for writing control software. The fuzzy controller and the neural network controller are implemented using a VHDL programming language, such as one or more combinations of fuzzy operations and rule bases in fuzzy logic, and one or more combinations of various neurons and network connections in a neural network model.
In the preferred scheme, the fuzzy controller and the neural network controller are connected in series to realize the hybrid controller, and the outputs of the fuzzy controller and the neural network controller are fused in a weighting, bias adjusting and other modes, for example, a DSP module in the FPGA hardware can realize the control algorithms with high speed and low power consumption. And performing joint debugging and debugging on the controller, and optimizing and adjusting parameters and working ranges of the controller according to actual conditions so as to achieve the optimal control effect. The established control system is applied to the seasoning preparation process, the physical and chemical properties of the seasoning are monitored in real time, and the proportion and stirring process of the seasoning are controlled according to a control strategy, so that the automatic and refined preparation process is realized.
Obtaining chemical characteristic data of various condiments, such as ingredient composition, physicochemical properties, taste sensory evaluation and the like, through literature investigation, laboratory test and the like; based on the collected chemical characteristic data, a prediction model is established by using a machine learning method or a deep learning method, and the common methods include linear regression, a support vector machine, a random forest, a neural network and the like, and a predicted taste evaluation result is output by inputting the chemical characteristic data of the seasoning.
In a preferred embodiment, a feedback system is built to monitor the chemical characteristics of the various flavors during the brewing process in real time and to input these data into a predictive model for taste prediction. Meanwhile, according to the difference between the predicted result and the target taste value, the addition amount and the proportion of the seasoning are controlled so as to achieve an accurate modulation effect.
Factors such as the nature of the condiment, the brewing process, etc. determine key parameters such as control period, target taste value, control algorithm, etc. when designing the feedback control system.
Optimizing contents including the adding sequence, time, temperature, pH value and the like of the seasonings according to the historical data obtained by the modulation result, and simultaneously monitoring and processing abnormal conditions in the modulation process.
In the preferred scheme, the pictures of different kinds of vegetables are collected, factors such as different angles, illumination and background are included, quality and diversity of the pictures are guaranteed, and different conditions can be identified by the model. And marking the collected vegetable pictures, and marking the vegetable types corresponding to each picture. Preprocessing the marked data, including data cleaning, data enhancement and other operations. A Convolutional Neural Network (CNN) based model, such as res net, VGG, etc., is selected for training. Training the model by using the marked data set, and performing parameter tuning in the training process. The accuracy of the model is evaluated by using indexes such as accuracy rate, recall rate and the like, and deployment is performed by using a prediction API or a local calling mode. And shooting vegetables in the designated area by using a sensor (such as a camera), inputting the shot pictures into a trained model for identification, and obtaining the types and related information of the vegetables.
In a preferred scheme, an execution model (such as sequence, shape to be processed by food materials and the like) is made from the trained dishes to derive an execution instruction, and a subsequent making step is further completed by an automatic flavoring machine and making system. ( Generating a manufacturing instruction: deriving execution instructions from a dish-making model )
The following are specific embodiments of the present invention:
embodiment one:
the old, 70 years old, has no fixation work, and the young is engaged in heavy physical labor, and the body has various basic diseases, especially with hypertension and gastropathy. The system was tried after the proposal of the family, at this time, 10:00 am, and the lunch time was 2 hours.
First, personal information is entered. The basic file of the health of the user (such as the old) is recorded, wherein the file key comprises dietary data such as the disease type, the severity of the disease, known food allergy or adverse reaction, drug treatment conditions and the like of the user, and data items with risks and relativity in the health physical examination data of the old are extracted and stored as a database 2.
Second, intelligent meal suggestion recommendation. According to the physical information of the user, the physical condition and the nutrition information of the user are input into a machine learning algorithm, and a matching model is established to determine which foods and dishes are most suitable for the user. The model may be adjusted based on factors such as the user's health goals, taste preferences, eating habits, disease type, etc.
Targeted dietary advice is provided based on the user's personal conditions (e.g., preferences and dietary restrictions), such as providing recipes, cooking advice, and health meal selection advice, among others.
In the preferred scheme, the food records of the old are input into the circulating neural network as sequence data, and the food habits and the nutrition states of the old are predicted through the training network, so that the nutrition conditions of the old are better judged, and corresponding nutrition intervention schemes are provided.
The food history of the old is used as an input sequence, and information such as salt, sugar, fat, nutrients, food additives and the like taken every day, and information such as age, height, weight, medical history and the like of the old are used as additional information. Through training the network, the nutritional status of the elderly can be predicted, and corresponding nutritional intervention schemes are provided, such as recommending low-salt, low-sugar and low-fat foods, limiting the intake of food additives for the elderly, and the like.
The old finishes the issuing of the instruction in the recommended dishes or searches dishes by himself through the WeChat applet.
The old man can select one of the most suitable food materials in a plurality of economic and convenient ways, such as buying vegetables by a robot or picking vegetables by a wire or taking or delivering goods by a wire according to comprehensive analysis recommendation (considering time, habit, position and the like) by completing the following prompt content of the system through interactive equipment such as voice and the like suitable for ageing transformation, such as retrieving coordinates of available sources of nearby food materials by the system.
The old selects dishes which meet the own taste and are beneficial to controlling hypertension and stomach diseases, namely the purple rice and the bean curd soup.
The old sorts and puts the prepared food materials into an automatic flavoring machine and manufacturing system according to the prompt, and starts the system. The system automatically completes the production of dishes.
In particular, when the bean curd soup is prepared, the system considers the hypertension condition of the old, controls the intake of salt and the like, and the blood pressure of the old is obviously controlled and the physical quality of the old becomes better and better through the effect of daily accumulation and moon.
Example 2:
automated seasoning and manufacturing systems, for example:
assuming that the system has n raw materials and condiments, x is used for each 1 ,x 2 ,...,x n Representing the mass fraction of them,
the following mathematical model is established by mass conservation and material conservation:
y 1 =f 1 (x 1 ,x 2 ,...,x n )
y 2 =f 2 (x 1 ,x 2 ,...,x n )
……
y m =f m (x 1 ,x 2 ,...,x n )
wherein x is i w i,j Represents the mole number of the ith substance in the jth reaction, x i w′ i,j The number of moles of the i-th substance produced in the j-th reaction is represented by k.
And (3) inputting the mole number of each element into a neural network by collecting chemical characteristic data in the dishes, and outputting a taste prediction result of the dishes.
The above embodiments are merely preferred embodiments of the present invention, the protection scope of the present invention is not limited thereto, and any simple changes or equivalent substitutions of technical solutions that can be obviously obtained by those skilled in the art within the technical scope of the present invention disclosed in the present invention belong to the protection scope of the present invention.
Claims (8)
1. A deep learning-based seasoning intake determination method, comprising:
acquiring body data of a historical user and nutritional information data ingested on the same day;
predicting the total amount of each flavoring to be taken on the next day based on the body data of the historical user and the daily intake nutritional data; and taking the predicted total amount of each flavoring to be taken on the next day, the physical data of the historical user and the daily nutritional data as a training set;
training an LSTM model by using body data of a historical user in the training set and daily intake nutrition data, wherein when the total amount of each seasoning output by the LSTM model is the same as the predicted total amount of each seasoning to be taken in the next day, the LSTM model is trained;
the body data of the user and the daily intake nutrition information are formed into sequence data, one time step is taken as one day, each time step is taken as the input of the LSTM model, and the data in three days are taken as the context of the input data, so that the consumption of each flavoring needed to be taken by the user in the next day is output.
2. The deep learning based seasoning intake determination method of claim 1, wherein the user's body data comprises: age, sex, height, weight, blood pressure, blood sugar, type of disease of the user, severity of disease of the user, known food allergy or adverse reaction, and medical treatment condition.
3. The deep learning based seasoning intake determining method according to claim 1, wherein the daily intake nutrition information of the user is inputted into a matrix decomposition model, and foods and dishes most suitable for the user are recommended by the matrix decomposition model.
4. A deep learning based flavor intake determination system, comprising:
the data acquisition module is used for acquiring physical data of a user and nutritional information data ingested on the same day;
the main control module is used for:
predicting the total amount of each flavoring to be taken on the next day based on the body data of the historical user and the daily intake nutritional data; and taking the predicted total amount of each flavoring to be taken on the next day, the physical data of the historical user and the daily nutritional data as a training set;
training an LSTM model by using body data of a historical user in the training set and daily intake nutrition data, wherein when the total amount of each seasoning output by the LSTM model is the same as the predicted total amount of each seasoning to be taken in the next day, the LSTM model is trained;
the body data of the user and the daily intake nutrition information are formed into sequence data, one time step is taken as one day, each time step is taken as the input of the LSTM model, and the data in three days are taken as the context of the input data, so that the consumption of each flavoring needed to be taken by the user in the next day is output.
5. A deep learning based flavoring intake determination system according to claim 3, further comprising:
the real-time monitoring module is used for monitoring chemical characteristic data of dishes in real time in the dish manufacturing process;
the main control module inputs the real-time chemical characteristic data of the dishes into a trained mlp neural network according to the real-time chemical characteristic data of the dishes, the mlp neural network model predicts the tastes of the dishes in real time, and the addition amount and the proportion of the seasonings are controlled according to the difference between the predicted result of the tastes of the dishes and the target taste value.
6. The deep learning based seasoning intake determination system of claim 5 wherein the real time monitoring module comprises:
the pH sensor is used for detecting the concentration of hydrogen ions in dishes;
the salinity sensor is used for detecting the concentration of salt ions in dishes;
the ammonia sensor is used for monitoring the ammonia concentration in dishes.
7. A deep learning based flavoring intake determination system according to claim 3, further comprising: a dish preparation module, comprising:
the inner container is used for cooking dishes or soup.
8. A deep learning based flavoring intake determination system according to claim 3, characterized in that the main control module comprises: and the fuzzy controller and the neural network controller are connected in series to form a hybrid controller.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310817345.5A CN117153338A (en) | 2023-07-05 | 2023-07-05 | Deep learning-based seasoning intake determining method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310817345.5A CN117153338A (en) | 2023-07-05 | 2023-07-05 | Deep learning-based seasoning intake determining method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117153338A true CN117153338A (en) | 2023-12-01 |
Family
ID=88901561
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310817345.5A Pending CN117153338A (en) | 2023-07-05 | 2023-07-05 | Deep learning-based seasoning intake determining method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117153338A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117577270A (en) * | 2024-01-15 | 2024-02-20 | 吉林大学 | Patient intelligent nutrition management method and system |
-
2023
- 2023-07-05 CN CN202310817345.5A patent/CN117153338A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117577270A (en) * | 2024-01-15 | 2024-02-20 | 吉林大学 | Patient intelligent nutrition management method and system |
CN117577270B (en) * | 2024-01-15 | 2024-04-26 | 吉林大学 | Patient intelligent nutrition management method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104303185B (en) | System and method for user's particular adjustments of nutrients intake | |
CN110289079A (en) | Wired home recipe supplying system, the method and device for pushing recipe | |
US9528972B2 (en) | Dynamic recipe control | |
Lutz et al. | Nutrition and diet therapy | |
US20160217420A1 (en) | Conditioning system for nutritional substances | |
US20130269543A1 (en) | System for Managing the Nutritional Content for Nutritional Substances | |
CN111261260B (en) | Diet recommendation system | |
US20140052722A1 (en) | Optimization-based regimen method and system for personalized diabetes and diet management | |
CN109243580A (en) | A kind of hypoglycemic intelligent nutrition catering system and dietary management method | |
KR20190009405A (en) | Automation Method for supplying customized menu | |
CN117153338A (en) | Deep learning-based seasoning intake determining method and system | |
CN111554379A (en) | Healthy food recipe recommendation method and device and computer readable storage medium | |
CN111161838A (en) | Reasonable diet method suitable for students, expert system and client | |
CN109461492A (en) | It is a kind of for alleviating the intelligent pantry and dietary management method of gout | |
CN116453652A (en) | Diabetes patient food intake control method, device, computer equipment and storage medium thereof | |
CN109509536A (en) | A kind of intelligent nutrition catering system and dietary management method for postpartum reparation | |
KR20180073779A (en) | Automation Method for supplying customized menu | |
CN113744840A (en) | System and method for realizing personalized instant nutrition evaluation and guidance based on centralized meal supply environment | |
WO2015195575A1 (en) | Dynamic recipe control | |
KR20040012394A (en) | The method and system for offering information to improve eating habits via internet | |
CN112116987A (en) | Meal collocation method and system based on multi-attribute decision | |
Mudry | Quantifying an American eater: Early USDA food guidance, and a language of numbers | |
CN115455065A (en) | Food material selecting and processing method and device | |
Falomir | Including hard restrictions into Diet4You menu planner | |
CN111435610A (en) | Method and device for recommending food and cooking appliance |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |