Cooking state recognition, model training method, system, equipment and storage medium

Abstract

The present invention discloses a cooking state recognition and model training method, system, device, and storage medium. The training method comprises: obtaining cooking data from at least two types of detection sensors under different cooking states during a historical cooking process; extracting multiple feature data from the cooking data; performing dimensionality reduction processing on the multiple feature data to obtain a preset number of reduced-dimensionality feature data; inputting the reduced-dimensionality feature data into a classification algorithm for training to obtain a cooking state recognition model; the cooking state recognition model uses the cooking data as input and the cooking state as output. The method extracts multiple feature data corresponding to the cooking data from the at least two types of detection sensors under different cooking states during the previously recorded cooking process; then performs dimensionality reduction processing using principal component analysis to extract and retain several reduced-dimensionality features with high information content; and then inputs the reduced-dimensionality features into a classification model for training to obtain a cooking mode prediction model.

Description

Technical Field

[0001] The present invention belongs to the field of cooking control, and in particular relates to a cooking state recognition, model training method, system, equipment and storage medium. Background Art

[0002] During the cooking process, the user's cooking movements and cooking mode information are particularly important for the intelligent control of the cooktop. Knowing the cooking mode information allows for intelligent adjustments to the cooking temperature, cooking time, and cooking power based on the corresponding mode, thereby improving cooking quality and user experience.

[0003] However, there are few existing solutions for cooking mode recognition, and they are relatively simple. They basically judge based on temperature values ​​and temperature changes. However, for different pots and pans, under the same fire conditions, due to differences in the thermal conductivity of their materials, their temperature change curves under the same cooking mode are also different. Therefore, judging the cooking mode only by temperature value changes will cause certain misjudgments. In other words, the single information obtained leads to a serious reduction in judgment accuracy, thus affecting the experience. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the defect in the prior art that the cooking mode recognition method is too single and leads to low judgment accuracy, and to provide a cooking state recognition, model training method, system, equipment and storage medium.

[0005] The present invention solves the above technical problems through the following technical solutions:

[0006] A cooking state recognition model training method, the training method comprising:

[0007] Acquire cooking data of at least two types of detection sensors under different cooking states during a historical cooking process;

[0008] extracting a plurality of feature data from the cooking data;

[0009] Performing dimensionality reduction processing on the plurality of feature data to obtain a preset number of reduced-dimensionality feature data;

[0010] Inputting the dimension-reduced feature data into a classification algorithm for training to obtain a cooking state recognition model;

[0011] The cooking state recognition model takes cooking data as input and takes cooking state as output.

[0012] Preferably, the types of detection sensors include temperature sensors, fire gear detection modules, pot material detection modules, pot weight detection sensors, and oil smoke concentration detection sensors;

[0013] The step of obtaining cooking data of at least two types of detection sensors under different cooking states during the historical cooking process specifically includes:

[0014] At least two of the temperature data, fire level data, cookware material data, cookware weight data and oil smoke concentration under different cooking conditions are obtained as the cooking data.

[0015] Preferably, if the cooking data includes temperature data, the step of extracting a plurality of characteristic data from the cooking data specifically includes:

[0016] Calculate at least one of the maximum temperature, the average temperature, the standard deviation of the temperature, the variance of the temperature, and the maximum positive slope of the temperature within the detection period as characteristic data according to the temperature data;

[0017] and / or,

[0018] If the cooking data includes fire level data, the step of extracting a plurality of feature data from the cooking data specifically includes:

[0019] Calculate at least one of the initial firepower gear, the longest-lasting firepower gear, and the average firepower gear within the detection period as feature data based on the firepower gear data;

[0020] and / or,

[0021] If the cooking data includes cookware material data, the step of extracting a plurality of feature data from the cooking data specifically includes:

[0022] At least one of thermal conductivity and specific heat capacity within a detection period is calculated based on the cookware material data as characteristic data.

[0023] Preferably, the step of performing dimensionality reduction processing on the plurality of feature data to obtain a preset number of dimensionality-reduced feature data specifically includes:

[0024] The plurality of feature data are subjected to dimensionality reduction processing based on a principal component analysis method, and a preset number of feature data ranked high in terms of information quantity are extracted as dimensionality reduction feature data.

[0025] Preferably, the cooking state includes any one of dry-roasting cooking, stewing cooking, frying cooking, and deep-frying cooking.

[0026] Preferably, the classification algorithm includes any one of a Kmeans algorithm, a SVM algorithm, and a linear regression algorithm.

[0027] A cooking state recognition model training system, the training system comprising:

[0028] A historical cooking data acquisition module, used to acquire cooking data from at least two types of detection sensors under different cooking states during a historical cooking process;

[0029] A feature data extraction module, configured to extract a plurality of feature data from the cooking data;

[0030] A dimensionality reduction module, configured to perform dimensionality reduction processing on the plurality of feature data to obtain a preset number of reduced-dimensionality feature data;

[0031] A training module, configured to input the dimension-reduced feature data into a classification algorithm for training, thereby obtaining a cooking state recognition model;

[0032] The cooking state recognition model takes cooking data as input and takes cooking state as output.

[0033] Preferably, the types of detection sensors include temperature sensors, fire gear detection modules, pot material detection modules, pot weight detection sensors, and oil smoke concentration detection sensors;

[0034] The historical cooking data acquisition module is specifically used for:

[0035] At least two of the temperature data, fire level data, cookware material data, cookware weight data and oil smoke concentration under different cooking conditions are obtained as the cooking data.

[0036] Preferably, if the cooking data includes temperature data, the feature data extraction module is specifically configured to:

[0037] Calculate at least one of the maximum temperature, the average temperature, the standard deviation of the temperature, the variance of the temperature, and the maximum positive slope of the temperature within the detection period as characteristic data according to the temperature data;

[0038] and / or,

[0039] If the cooking data includes fire power level data, the feature data extraction module is specifically used to:

[0040] Calculate at least one of the initial firepower gear, the longest-lasting firepower gear, and the average firepower gear within the detection period as feature data based on the firepower gear data;

[0041] and / or,

[0042] If the cooking data includes cookware material data, the feature data extraction module is specifically configured to:

[0043] At least one of thermal conductivity and specific heat capacity within a detection period is calculated based on the cookware material data as characteristic data.

[0044] Preferably, the dimensionality reduction module is specifically used to:

[0045] The plurality of feature data are subjected to dimensionality reduction processing based on a principal component analysis method, and a preset number of feature data ranked high in terms of information quantity are extracted as dimensionality reduction feature data.

[0046] Preferably, the cooking state includes any one of dry-roasting cooking, stewing cooking, frying cooking, and deep-frying cooking.

[0047] Preferably, the classification algorithm includes any one of a Kmeans algorithm, a SVM algorithm, and a linear regression algorithm.

[0048] A cooking state recognition method, the recognition method comprising:

[0049] acquiring real-time cooking data from at least two types of detection sensors during the cooking process;

[0050] Inputting the real-time cooking data into the cooking state recognition model trained by the above-mentioned cooking state recognition model training method;

[0051] Output real-time cooking status.

[0052] A cooking state recognition system, comprising:

[0053] A real-time data acquisition module, used to acquire real-time cooking data from at least two types of detection sensors during the cooking process;

[0054] an input module, configured to input the real-time cooking data into the cooking state recognition model trained by the above-mentioned cooking state recognition model training method;

[0055] Output module, used to output real-time cooking status.

[0056] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the processor implements the above-mentioned cooking state recognition model training method or the above-mentioned cooking state recognition method.

[0057] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned cooking state recognition model training method or the above-mentioned cooking state recognition method.

[0058] The positive progressive effect of the present invention is that: the present invention extracts multiple feature data corresponding to the cooking data of at least two types of detection sensors through multi-dimensional perception information under different cooking states recorded in advance, and then uses the principal component analysis method for dimensionality reduction processing to extract and retain several reduced-dimensional features with large information content, and then inputs the reduced-dimensional features into the classification model for training to obtain a cooking mode prediction model. Furthermore, in the actual cooking process, the real-time collected data is input into the trained model to directly output the corresponding cooking mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flowchart of the cooking state recognition model training method according to embodiment 1 of the present invention.

[0060] Figure 2 This is a module diagram of the cooking state recognition model training system according to embodiment 2 of the present invention.

[0061] Figure 3 This is a flow chart of the cooking status identification method according to embodiment 3 of the present invention.

[0062] Figure 4 This is a module diagram of a cooking status recognition system according to embodiment 4 of the present invention.

[0063] Figure 5 This is a schematic structural diagram of an electronic device according to embodiment 5 of the present invention. DETAILED DESCRIPTION

[0064] The present invention is further described below by way of examples, but the present invention is not limited to the scope of the examples.

[0065] Example 1

[0066] A cooking state recognition model training method, such as Figure 1 As shown, the training method includes:

[0067] Step 101: Acquire cooking data of at least two types of detection sensors under different cooking states during a historical cooking process;

[0068] The cooking state includes any one of dry cooking, stewing, frying, and deep-frying. The original training dataset is obtained by collecting sensory information data from different cookware in different cooking modes. The types of cookware include but are not limited to stainless steel pots, cast iron pots, casseroles, glass pots, and other pots.

[0069] It should be noted that the settings of detection sensors and cooking status are not limited to the above-mentioned ones. Considering the feasibility, accuracy requirements and personalized customization needs, the classification results of cooking modes can also include: stir-frying, sautéing, deep-frying, cooking, frying, sticking, roasting, braising, stewing, steaming, blanching, boiling, stewing, sautéing, mixing, marinating, baking, braising, freezing, candied, honey-glazed, smoking, rolling, sliding, and baking.

[0070] In addition, the types of detection sensors include temperature sensors, fire gear detection modules, pot material detection modules, pot weight detection sensors, and oil smoke concentration detection sensors. Furthermore, step 101 specifically includes:

[0071] At least two of the following data, including temperature data, power level data, pot material data, pot weight data, and oil smoke concentration data, are obtained under different cooking conditions as cooking data. It should be noted that the type and number of detection sensors can be set as needed, also taking into account feasibility, accuracy requirements, and personalized customization requirements.

[0072] Step 102: extracting a plurality of feature data from the cooking data;

[0073] In this embodiment, if the cooking data includes temperature data, step 102 specifically includes:

[0074] At least one of the maximum temperature, the average temperature, the standard deviation of the temperature, the variance of the temperature, and the maximum positive slope of the temperature within the detection period is calculated according to the temperature data as characteristic data;

[0075] If the cooking data includes fire level data, step 102 specifically includes:

[0076] At least one of the initial firepower gear, the longest-lasting firepower gear, and the average firepower gear within the detection period is calculated as feature data according to the firepower gear data;

[0077] If the cooking data includes cookware material data, step 102 specifically includes:

[0078] At least one of thermal conductivity and specific heat capacity within a detection period is calculated based on the cookware material data as characteristic data.

[0079] It should be noted that the principal component analysis method is used to reduce the dimension of all the feature data, including multiple features corresponding to temperature, firepower and cookware, and retain several features for subsequent training. For example, the maximum temperature, average firepower level and thermal conductivity may be retained in the end.

[0080] Step 103: Perform dimensionality reduction processing on the plurality of feature data to obtain a preset number of reduced-dimensional feature data;

[0081] Wherein, step 103 specifically includes:

[0082] Based on the principal component analysis method, dimensionality reduction processing is performed on multiple feature data, and a preset number of feature data with the highest information quantity are extracted as dimensionality reduction feature data.

[0083] It should be noted that when dimensionality reduction is performed on all acquired feature data, the feature data extracted from multi-dimensional perceptual information is multi-dimensional and contains a lot of redundant information. In this case, the eigenvalues ​​of too many dimensions will drown out the useful information, preventing the desired effect. Therefore, principal component analysis (PCA) is considered to perform dimensionality reduction projection on all feature data, thereby filtering out more useful feature data. The principle of PCA is to project the original high-dimensional data into a low-dimensional space by finding a new vector basis, while maintaining the maximum variance within each dimension of the data. This removes noise with small variance and retains the principal components with high information content. Specifically, a PCA algorithm based on eigenvalue decomposition of the covariance matrix is ​​used to calculate the eigenvalues ​​and eigenvectors of the covariance matrix. The eigenvalues ​​are sorted from large to small, and the eigenvectors corresponding to the eigenvalues ​​with the largest absolute values ​​are selected as the data projection directions. The sample dataset is then dimensionalized in the feature space to obtain new principal component feature variables. The principal components with the highest information content are retained as the dimensionality reduction feature data for subsequent algorithm training.

[0084] Step 104: Input the dimension-reduced feature data into a classification algorithm for training to obtain a cooking state recognition model; wherein the classification algorithm includes any one of a Kmeans algorithm, a SVM algorithm, and a linear regression algorithm.

[0085] The cooking state recognition model takes cooking data as input and cooking state as output.

[0086] In this embodiment, the multi-dimensional perception information under different cooking states in the cooking process is recorded in advance to extract multiple feature data corresponding to the cooking data of at least two types of detection sensors. The principal component analysis method is then used for dimensionality reduction processing to extract and retain several reduced-dimensional features with large information content. The reduced-dimensional features are then input into the classification model for training to obtain a cooking mode prediction model.

[0087] Example 2

[0088] A cooking state recognition model training system, such as Figure 2 As shown, the training system includes:

[0089] A historical cooking data acquisition module 11 is used to acquire cooking data from at least two types of detection sensors under different cooking states during a historical cooking process;

[0090] The cooking state includes any one of dry cooking, stewing, stir-frying, and deep-frying. The original training dataset is obtained by collecting sensory information data from different cookware in different cooking modes. The types of cookware include, but are not limited to, stainless steel pots, cast iron pots, casseroles, and glass pots.

[0091] It should be noted that the settings of detection sensors and cooking status are not limited to the above-mentioned ones. Considering the feasibility, accuracy requirements and personalized customization needs, the classification results of cooking modes can also include: stir-frying, sautéing, deep-frying, cooking, frying, sticking, roasting, braising, stewing, steaming, blanching, boiling, stewing, sautéing, mixing, marinating, baking, braising, freezing, candied, honey-glazed, smoking, rolling, sliding, and baking.

[0092] In addition, the categories of the detection sensors include temperature sensors, fire gear detection modules, pot material detection modules, pot weight detection sensors, and oil smoke concentration detection sensors. Furthermore, the historical cooking data acquisition module 11 is specifically used to:

[0093] At least two of the temperature data, fire level data, cookware material data, cookware weight data and oil smoke concentration under different cooking conditions are obtained as the cooking data.

[0094] It should be noted that, also considering feasibility, accuracy requirements, and personalized customization needs, the type and number of detection sensors can be set as needed.

[0095] A feature data extraction module 12 is configured to extract a plurality of feature data from the cooking data;

[0096] In this embodiment, if the cooking data includes temperature data, the feature data extraction module 12 is specifically configured to:

[0097] Calculate at least one of the maximum temperature, the average temperature, the standard deviation of the temperature, the variance of the temperature, and the maximum positive slope of the temperature within the detection period as characteristic data according to the temperature data;

[0098] If the cooking data includes fire power level data, the feature data extraction module 12 is specifically configured to:

[0099] Calculate at least one of the initial firepower gear, the longest-lasting firepower gear, and the average firepower gear within the detection period as feature data based on the firepower gear data;

[0100] If the cooking data includes cookware material data, the feature data extraction module 12 is specifically configured to:

[0101] At least one of thermal conductivity and specific heat capacity within a detection period is calculated based on the cookware material data as characteristic data.

[0102] It should be noted that the principal component analysis method is used to reduce the dimension of all the feature data, including multiple features corresponding to temperature, firepower and cookware, and retain several features for subsequent training. For example, the maximum temperature, average firepower level and thermal conductivity may be retained in the end.

[0103] A dimensionality reduction module 13 is configured to perform dimensionality reduction processing on the plurality of feature data to obtain a preset number of reduced-dimensionality feature data;

[0104] The dimension reduction module 13 is specifically used for:

[0105] The plurality of feature data are subjected to dimensionality reduction processing based on a principal component analysis method, and a preset number of feature data ranked high in information quantity are extracted as dimensionality reduction feature data.

[0106] It should be noted that when dimensionality reduction is performed on all acquired feature data, the feature data extracted from multi-dimensional perceptual information is multi-dimensional and contains a lot of redundant information. In this case, the eigenvalues ​​of too many dimensions will drown out the useful information, preventing the desired effect. Therefore, principal component analysis (PCA) is considered to perform dimensionality reduction projection on all feature data, thereby filtering out more useful feature data. The principle of PCA is to project the original high-dimensional data into a low-dimensional space by finding a new vector basis, while maintaining the maximum variance within each dimension of the data. This removes noise with small variance and retains the principal components with high information content. Specifically, a PCA algorithm based on eigenvalue decomposition of the covariance matrix is ​​used to calculate the eigenvalues ​​and eigenvectors of the covariance matrix. The eigenvalues ​​are sorted from large to small, and the eigenvectors corresponding to the eigenvalues ​​with the largest absolute values ​​are selected as the data projection directions. The sample dataset is then dimensionalized in the feature space to obtain new principal component feature variables. The principal components with the highest information content are retained as the dimensionality reduction feature data for subsequent algorithm training.

[0107] The training module 14 is used to input the dimension-reduced feature data into a classification algorithm for training to obtain a cooking state recognition model; wherein the classification algorithm includes any one of a Kmeans algorithm, a SVM algorithm, and a linear regression algorithm.

[0108] The cooking state recognition model takes cooking data as input and takes cooking state as output.

[0109] In this embodiment, the multi-dimensional perception information under different cooking states in the cooking process is recorded in advance to extract multiple feature data corresponding to the cooking data of at least two types of detection sensors. The principal component analysis method is then used for dimensionality reduction processing to extract and retain several reduced-dimensional features with large information content. The reduced-dimensional features are then input into the classification model for training to obtain a cooking mode prediction model.

[0110] Example 3

[0111] A cooking state recognition method, such as Figure 3 As shown, the identification method includes:

[0112] Step 201: Acquire real-time cooking data from at least two types of detection sensors during a cooking process;

[0113] It should be noted that the data of each sensor are not limited to being collected at the same time. The cooking status can be identified according to a set period, such as once every 1 minute, and the collection period of the detection sensor, for example, temperature is collected once every 1 second, gear position is collected once every 5 seconds, and material is collected once every 5 minutes. When extracting real-time cooking data, the latest data of each sensor can be used as input data.

[0114] Step 202: input the real-time cooking data into the cooking state recognition model trained by the above-mentioned cooking state recognition model training method;

[0115] Step 203: Output the real-time cooking status.

[0116] In this embodiment, during the actual cooking process, the detection sensor collects cooking data in real time, and then directly inputs the cooking data into the cooking state recognition model trained using the cooking state recognition model training method of Example 1, and can directly output the current cooking state. In addition, depending on the classification algorithm selected during the cooking state recognition model training process, the classification category and classification probability can also be output as needed.

[0117] Example 4

[0118] A cooking status recognition system, such as Figure 4 As shown, the system includes:

[0119] A real-time data acquisition module 21 is used to acquire real-time cooking data from at least two types of detection sensors during the cooking process;

[0120] It should be noted that the data of each sensor are not limited to being collected at the same time. The cooking status can be identified according to a set period, such as once every 1 minute, and the collection period of the detection sensor, for example, temperature is collected once every 1 second, gear position is collected once every 5 seconds, and material is collected once every 5 minutes. When extracting real-time cooking data, the latest data of each sensor can be used as input data.

[0121] An input module 22, configured to input the real-time cooking data into the cooking state recognition model trained by the above-mentioned cooking state recognition model training method;

[0122] The output module 23 is used to output the real-time cooking status.

[0123] In this embodiment, during the actual cooking process, the detection sensor collects cooking data in real time, and then directly inputs the cooking data into the cooking state recognition model trained using the cooking state recognition model training method of Example 1, and can directly output the current cooking state. In addition, depending on the classification algorithm selected during the cooking state recognition model training process, the classification category and classification probability can also be output as needed.

[0124] Example 5

[0125] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the cooking state recognition model training method described in Example 1 or the cooking state recognition method described in Example 3 is implemented.

[0126] Figure 5 This is a schematic structural diagram of an electronic device provided in this embodiment. Figure 5 A block diagram of an exemplary electronic device 90 suitable for use in implementing embodiments of the present invention is shown. Figure 5 The electronic device 90 shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0127] like Figure 5 As shown, the electronic device 90 may be a general-purpose computing device, such as a server device. Components of the electronic device 90 may include, but are not limited to, at least one processor 91, at least one memory 92, and a bus 93 connecting different system components (including the memory 92 and the processor 91).

[0128] The bus 93 includes a data bus, an address bus, and a control bus.

[0129] The memory 92 may include a volatile memory, such as a random access memory (RAM) 921 and / or a cache memory 922 , and may further include a read-only memory (ROM) 923 .

[0130] The memory 92 may also include a program tool 925 having a set (at least one) of program modules 924, such program modules 924 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.

[0131] The processor 91 executes various functional applications and data processing by running computer programs stored in the memory 92 .

[0132] The electronic device 90 can also communicate with one or more external devices 94 (e.g., a keyboard, pointing device, etc.). Such communication can occur via an input / output (I / O) interface 95. Furthermore, the electronic device 90 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 96. The network adapter 96 communicates with other modules of the electronic device 90 via a bus 93. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with the electronic device 90, including but not limited to microcode, device drivers, redundant processors, external disk drive arrays, RAID (RAID) systems, tape drives, and data backup storage systems.

[0133] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, depending on the embodiment of the present application, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0134] Example 6

[0135] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the cooking state recognition model training method described in Example 1, or implements the cooking state recognition method described in Example 3. The readable storage medium may more specifically include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0136] In a possible embodiment, the present invention can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the cooking state recognition model training method described in Example 1, or implement the cooking state recognition method described in Example 3.

[0137] The program code for executing the present invention may be written in any combination of one or more programming languages, and may be executed entirely on the user device, partially on the user device, as an independent software package, partially on the user device and partially on a remote device, or entirely on the remote device.

[0138] Although specific embodiments of the present invention have been described above, those skilled in the art will appreciate that these are merely illustrative and that the scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, and such changes and modifications are intended to fall within the scope of the present invention.

Claims

1. A cooking state recognition model training method, characterized in that: The training method comprises: Acquire cooking data of at least two types of detection sensors under different cooking states during a historical cooking process; extracting a plurality of feature data from the cooking data; Performing dimensionality reduction processing on the plurality of feature data to obtain a preset number of reduced-dimensionality feature data; Inputting the dimension-reduced feature data into a classification algorithm for training to obtain a cooking state recognition model; The cooking state recognition model takes cooking data as input and takes cooking state as output; The categories of the detection sensors include temperature sensors, fire gear detection modules, pot material detection modules, pot weight detection sensors, and oil smoke concentration detection sensors; The step of obtaining cooking data of at least two types of detection sensors under different cooking states during the historical cooking process specifically includes: acquiring at least two of temperature data, fire level data, pot material data, pot weight data, and oil smoke concentration under different cooking conditions as the cooking data; If the cooking data includes temperature data, the step of extracting a plurality of characteristic data from the cooking data specifically includes: Calculate at least one of the maximum temperature, the average temperature, the standard deviation of the temperature, the variance of the temperature, and the maximum positive slope of the temperature within the detection period as characteristic data according to the temperature data; and, If the cooking data includes fire level data, the step of extracting a plurality of feature data from the cooking data specifically includes: Calculate at least one of the initial firepower gear, the longest-lasting firepower gear, and the average firepower gear within the detection period as feature data based on the firepower gear data; and, If the cooking data includes cookware material data, the step of extracting a plurality of feature data from the cooking data specifically includes: At least one of thermal conductivity and specific heat capacity within a detection period is calculated based on the cookware material data as characteristic data.

2. The cooking state recognition model training method according to claim 1, wherein: The step of performing dimensionality reduction processing on the plurality of feature data to obtain a preset number of dimensionality-reduced feature data specifically includes: The plurality of feature data are subjected to dimensionality reduction processing based on a principal component analysis method, and a preset number of feature data ranked high in terms of information quantity are extracted as dimensionality reduction feature data.

3. The cooking state recognition model training method according to claim 1, wherein: The cooking state includes any one of dry-roasting cooking, stewing cooking, frying cooking, and deep-frying cooking.

4. The cooking state recognition model training method according to claim 1, wherein: The classification algorithm includes any one of a Kmeans algorithm, a SVM algorithm, and a linear regression algorithm.

5. A cooking status recognition method, characterized in that: The identification method comprises: acquiring real-time cooking data from at least two types of detection sensors during the cooking process; inputting the real-time cooking data into a cooking state recognition model trained by the cooking state recognition model training method according to any one of claims 1 to 4; Output real-time cooking status.

6. A cooking status recognition system, characterized in that: The system comprises: A real-time data acquisition module, used to acquire real-time cooking data from at least two types of detection sensors during the cooking process; An input module, configured to input the real-time cooking data into a cooking state recognition model trained by the cooking state recognition model training method according to any one of claims 1 to 4; Output module, used to output real-time cooking status.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the cooking state recognition model training method according to any one of claims 1 to 4 is implemented, or the cooking state recognition method according to claim 5 is implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the cooking state recognition model training method according to any one of claims 1 to 4 is implemented, or the cooking state recognition method according to claim 5 is implemented.

Images