Heat preservation meal furnace operation control method and system based on artificial intelligence

Abstract

The invention provides a heat preservation meal furnace operation control method and system based on artificial intelligence, and the method comprises the steps: controlling a preset camera module to shoot the food on a dinner plate after the food is placed on the dinner plate of a heat preservation meal furnace, and obtaining a first food image; receiving a first food image sent by the camera module, and preprocessing the first food image to obtain a second food image; and calling a pre-trained food image analysis model to analyze the second food image, identifying to obtain a target dish type to which the food in the second food image belongs, determining the optimal heat preservation temperature of the target dish type, and setting the heat preservation temperature of the heat preservation meal furnace as the optimal heat preservation temperature. Automatic and intelligent control over the temperature of the heat preservation meal furnace is achieved in an artificial intelligence mode, manual intervention is not needed in the whole process, and operation convenience and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of heat preservation dining stoves, and in particular to an operation control method and system of a heat preservation dining stove based on artificial intelligence. Background Art

[0002] A heat preservation oven is a device used to keep food at a set temperature. It is usually used in the catering service industry, such as schools, hospitals, corporate canteens, fast food restaurants, etc., to ensure that food remains at a suitable serving temperature for a long time.

[0003] Insulated meal ovens usually use electric heating elements to provide heat, and monitor the temperature inside the oven through temperature sensors. When the temperature is lower than the set temperature, the heating element starts working to heat the food to the set temperature; when the temperature reaches or exceeds the set temperature, the heating element stops working to maintain the temperature of the food.

[0004] As we all know, different types of dishes have their optimal insulation temperatures to maintain their flavor, taste and nutrition. For example, the appropriate insulation temperatures for hot dishes and cold dishes are significantly different. The current method for setting the optimal insulation temperature for different types of dishes is to construct a comparison table of different types of dishes and the optimal insulation temperature. After the staff puts the food into the plate of the insulation dining furnace, they determine the optimal insulation temperature of the dish type based on the comparison table, and manually set the insulation temperature of the insulation dining furnace to the optimal insulation temperature. However, the comparison table is difficult to cover all types of dishes, especially some special or uncommon dishes; at the same time, the method requires manual identification of the dish type and search for the comparison table, which is a process prone to errors, especially in a busy catering service environment, and the manual setting of the optimal insulation temperature is also low in efficiency and intelligence. Therefore, the current insulation dining furnace cannot accurately, efficiently and intelligently set the optimal insulation temperature of the dish type.

[0005] For example, the current patent application No. 202311663276.3 discloses a temperature-controlled heat preservation dining stove control system based on artificial intelligence, in which the temperature analysis module is provided with a comparison table of dishes and target heat preservation temperature ranges, and compares the lower layer temperature of the dish with the lowest temperature of the target heat preservation temperature range to determine the insulation type of the dish, including cooling and heating, to ensure that the food is kept at the most suitable temperature. However, the technical solution uses a comparison table to cover all types of dishes, and requires manual identification of dish types and search for comparison tables, which is cumbersome and error-prone. Summary of the invention

[0006] The present invention provides an artificial intelligence-based heat preservation dining furnace operation control method and system, so as to accurately, efficiently and intelligently set the optimal heat preservation temperature of food in the heat preservation dining furnace.

[0007] To solve the above problems, the present invention adopts the following technical solutions:

[0008] The present invention provides an artificial intelligence-based heat preservation dining stove operation control method, comprising:

[0009] When food is placed on a dinner plate of the heat preservation dining oven, a preset camera module is controlled to shoot the food on the dinner plate to obtain a first food image;

[0010] receiving a first food image sent by the camera module, and preprocessing the first food image to obtain a second food image;

[0011] Calling a pre-trained food image analysis model to analyze the second food image, identifying a target dish type to which the food in the second food image belongs and determining an optimal heat preservation temperature for the target dish type, wherein the food image analysis model is a convolutional neural network, and is used to identify the dish type in the food image and determine the optimal heat preservation temperature for the dish type;

[0012] The insulation temperature of the insulation dining stove is set to the optimal insulation temperature.

[0013] Furthermore, before calling the pre-trained food image analysis model to analyze the second food image, identifying the target dish type to which the food in the second food image belongs and determining the optimal insulation temperature of the target dish type, the method further includes:

[0014] Collecting image datasets of different dish types, wherein the image datasets include historical food images of multiple dish types and the optimal keeping temperature of each dish type;

[0015] Preprocessing the image data set to obtain a standard image data set;

[0016] Using the standard image data set to train a convolutional neural network, and calculating a loss value of the trained convolutional neural network based on a preset loss function;

[0017] When the loss value is lower than a preset loss value, the trained convolutional neural network is used as a food image analysis model.

[0018] Preferably, the preprocessing of the image data set to obtain a standard image data set includes:

[0019] Performing format unification, size adjustment, color space conversion, standardization, and data enhancement processing on each historical food image in the image dataset to obtain a plurality of standard food images;

[0020] The plurality of standard food images are combined into a standard image dataset.

[0021] Furthermore, after calculating the loss value of the trained convolutional neural network based on a preset loss function, the method further includes:

[0022] When the loss value is not less than a preset loss value, based on the loss value, calculating a first gradient of a preset loss function with respect to a predicted output of the convolutional neural network, and a second gradient of an output layer activation function with respect to an input of the convolutional neural network;

[0023] After multiplying the first gradient by the second gradient, an output gradient of a preset loss function with respect to an output layer of the convolutional neural network is calculated;

[0024] Calculating the gradient of the output layer of the convolutional neural network with respect to the output of the hidden layer to obtain a third gradient;

[0025] Calculating the gradient of the hidden layer activation function of the convolutional neural network with respect to the input to obtain a fourth gradient;

[0026] After multiplying the output gradient, the third gradient and the fourth gradient, a target gradient is obtained;

[0027] After multiplying each model parameter of the convolutional neural network by the target gradient respectively, a plurality of target model parameters of the convolutional neural network are obtained;

[0028] Correspondingly updating the model parameters of the convolutional neural network based on the multiple target model parameters;

[0029] The convolutional neural network after updating the model parameters is trained again using the standard image data set until the loss value is lower than the preset loss value, thereby obtaining a trained food image analysis model.

[0030] Furthermore, after setting the insulation temperature of the insulation dining stove to the optimal insulation temperature, the method further comprises:

[0031] Setting a blending time for the food;

[0032] When the stirring time is reached, a preset stirring module is controlled to stir the food on the plate.

[0033] Furthermore, after setting the insulation temperature of the insulation dining stove to the optimal insulation temperature, the method further comprises:

[0034] controlling a preset camera module in real time to shoot the food on the dinner plate to obtain a third food image;

[0035] receiving the third food image sent by the camera module;

[0036] Calling the food image analysis model to analyze the third food image and generate an analysis result;

[0037] When it is determined according to the analysis result that there is no food on the dinner plate, a reminder to add food is sent to the terminal to which the staff belongs, and the heat preservation function of the heat preservation dinner stove is turned off.

[0038] Furthermore, after setting the insulation temperature of the insulation dining stove to the optimal insulation temperature, the method further comprises:

[0039] Acquiring temperature data of food inside the heat preservation dining stove;

[0040] Extracting characteristic information of the temperature data to obtain temperature characteristics, wherein the temperature characteristics include temperature deviation, deviation change rate and integrated absolute error;

[0041] Input the temperature characteristics and the optimal insulation temperature into a pre-trained neural network model, and output the optimal PID parameters;

[0042] The parameters of the PID controller of the heat-insulating dining stove are adjusted based on the optimal PID parameters, and the adjustment parameters and system response of the PID controller of the heat-insulating dining stove are recorded.

[0043] Preferably, calling a pre-trained food image analysis model to analyze the second food image, identifying the target dish type to which the food in the second food image belongs and determining the optimal insulation temperature of the target dish type includes:

[0044] Calling a pre-trained food image analysis model to extract multiple key features of the second food image;

[0045] identifying multiple types of ingredients and target cooking methods of the second food image based on multiple key features of the second food image;

[0046] Determining a target dish type to which the food in the second food image belongs based on the multiple types of ingredients and the target cooking methods;

[0047] When it is determined that the target dish type is a new dish type, calculating the pixel ratio of each type of food in the second food image to all types of food, and obtaining the weight of each type of food;

[0048] Querying the optimal keeping-warm temperature corresponding to each type of food under the target cooking method;

[0049] The optimal insulation temperature of each type of food is weighted and summed with the corresponding weight to obtain the optimal insulation temperature of the target dish type.

[0050] Furthermore, after determining the target dish type of the food in the second food image based on the multiple types of ingredients and the target cooking methods, the method further includes:

[0051] When it is determined that the target dish type is not a new dish type, the optimal insulation temperature of the target dish type is obtained from a database.

[0052] The present invention provides an artificial intelligence-based heat preservation dining stove operation control system, comprising:

[0053] A control module, configured to control a preset camera module to photograph the food on the dinner plate to obtain a first food image after the food is placed on the dinner plate of the heat preservation dining oven;

[0054] a preprocessing module, configured to receive the first food image sent by the camera module, and preprocess the first food image to obtain a second food image;

[0055] an analysis module, configured to call a pre-trained food image analysis model to analyze the second food image, identify a target dish type to which the food in the second food image belongs, and determine an optimal heat preservation temperature for the target dish type, wherein the food image analysis model is a convolutional neural network, configured to identify the dish type in the food image and determine an optimal heat preservation temperature for the dish type;

[0056] A setting module is used to set the insulation temperature of the insulation dining stove to the optimal insulation temperature.

[0057] Compared with the prior art, the technical solution of the present invention has at least the following advantages:

[0058] The artificial intelligence-based heat preservation dining furnace operation control method and system provided by the present invention receive a first food image sent by a camera module, pre-process the first food image to obtain a second food image, call a pre-trained food image analysis model to analyze the second food image, identify the target dish type to which the food in the second food image belongs and determine the optimal heat preservation temperature of the target dish type, set the heat preservation temperature of the heat preservation dining furnace to the optimal heat preservation temperature, and use a convolutional neural network to analyze the food image, so as to accurately identify the dish type and determine the optimal heat preservation temperature of the dish type, and automatically set the heat preservation temperature of the heat preservation dining furnace to the optimal heat preservation temperature, thereby realizing automatic and intelligent control of the temperature of the heat preservation dining furnace by artificial intelligence, and the whole process does not require manual intervention, thereby improving the convenience and efficiency of operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flowchart of an embodiment of the heat preservation dining stove operation control method based on artificial intelligence of the present invention;

[0060] Figure 2 This is a flowchart of another embodiment of the heat preservation dining stove operation control method based on artificial intelligence of the present invention;

[0061] Figure 3 It is a flowchart of another embodiment of the heat preservation dining stove operation control method based on artificial intelligence of the present invention;

[0062] Figure 4 The present invention is a structural block diagram of an embodiment of the heat preservation dining stove operation control system based on artificial intelligence. DETAILED DESCRIPTION

[0063] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0064] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The sequence numbers of the operations, such as S11, S12, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0065] It will be understood by those of ordinary skill in the art that, unless expressly stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0066] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as here.

[0067] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0068] Please refer to Figure 1 The present invention provides an artificial intelligence-based heat preservation dining stove operation control method, comprising the following steps:

[0069] S11, when food is placed on a dinner plate of the heat preservation dining oven, controlling a preset camera module to shoot the food on the dinner plate to obtain a first food image;

[0070] S12, receiving a first food image sent by the camera module, and preprocessing the first food image to obtain a second food image;

[0071] S13, calling a pre-trained food image analysis model to analyze the second food image, identifying the target dish type to which the food in the second food image belongs and determining the optimal insulation temperature of the target dish type, wherein the food image analysis model is a convolutional neural network, and is used to identify the dish type in the food image and determine the optimal insulation temperature of the dish type;

[0072] S14, setting the insulation temperature of the insulation dining stove to the optimal insulation temperature.

[0073] In this embodiment, a camera module can be integrated next to the heat preservation dining stove. When food is placed on the dinner plate of the heat preservation dining stove, the preset camera module is automatically controlled to shoot the food on the dinner plate to obtain a first food image. Or when the dinner plate is placed in the heat preservation dining stove, the camera module is automatically triggered to shoot the food on the dinner plate to obtain visual information of the food for subsequent image analysis. Among them, the position design of the camera module needs to consider the influence of water vapor generated by the food in the heat preservation dining stove to avoid blurring the captured image. For example, the camera module can be installed diagonally above the heat preservation dining stove to minimize the influence of water vapor in the food while ensuring that a complete image of the food is captured.

[0074] The food image captured by the camera module is received and preprocessed to form a second food image. The preprocessing method may include adjusting the image size, cropping, rotation correction, color correction, normalization, etc. to ensure that the image is suitable for input into the convolutional neural network, thereby improving the recognition accuracy and robustness of the model.

[0075] Then, the system calls the pre-trained food image analysis model to analyze the second food image, identify the target dish type to which the food in the second food image belongs, and determine the optimal insulation temperature of the target dish type. For example, a pre-trained convolutional neural network can be used to analyze the pre-processed second food image. The model can identify the dish type in the image and determine the optimal insulation temperature for each dish type based on the knowledge learned from the training data, thereby realizing intelligent identification of food and determination of the optimal insulation temperature.

[0076] Finally, according to the optimal insulation temperature determined by the food image analysis model, the temperature setting of the insulation dining furnace is automatically adjusted to ensure that the food can be kept at the optimal temperature, thereby realizing intelligent control of the insulation dining furnace and improving the insulation effect of food.

[0077] For example, suppose a cafeteria uses this technical solution to manage food in an insulated dining stove. When the staff puts the dining tray with dishes into the dining stove, the system controls the camera module to automatically shoot the food, obtain the image of the food, and ensure that the image quality is suitable for deep learning analysis. After the image quality is qualified, the food in the dining tray is identified as "braised pork" or "cold cucumber" based on the food image, and the optimal insulation temperature of the braised pork is determined to be 60°C and the optimal insulation temperature of the cold cucumber is 10°C based on the model training data. The insulated dining stove automatically adjusts the insulation temperature based on the recognition result, setting the insulation temperature of the braised pork area to 60°C and the insulation temperature of the cold cucumber area to 10°C, thereby reducing the need for manual temperature adjustment through automated temperature control, improving work efficiency, and ensuring that food is served in the best condition by precisely controlling the insulation temperature of each type of dish, so as to improve food quality and taste and enhance the dining experience of customers.

[0078] The artificial intelligence-based heat preservation dining furnace operation control method provided by the present invention receives a first food image sent by a camera module, pre-processes the first food image to obtain a second food image, calls a pre-trained food image analysis model to analyze the second food image, identifies the target dish type to which the food in the second food image belongs and determines the optimal heat preservation temperature of the target dish type, sets the heat preservation temperature of the heat preservation dining furnace to the optimal heat preservation temperature, and uses a convolutional neural network to analyze the food image, so as to accurately identify the dish type and determine the optimal heat preservation temperature of the dish type, and automatically sets the heat preservation temperature of the heat preservation dining furnace to the optimal heat preservation temperature, thereby realizing automatic and intelligent control of the temperature of the heat preservation dining furnace by artificial intelligence, and the whole process does not require manual intervention, thereby improving the convenience and efficiency of operation.

[0079] In one embodiment, before calling a pre-trained food image analysis model to analyze the second food image, identifying the target dish type to which the food in the second food image belongs and determining the optimal insulation temperature of the target dish type, the method further includes:

[0080] Collecting image datasets of different dish types, wherein the image datasets include historical food images of multiple dish types and the optimal keeping temperature of each dish type;

[0081] Preprocessing the image data set to obtain a standard image data set;

[0082] Using the standard image data set to train a convolutional neural network, and calculating a loss value of the trained convolutional neural network based on a preset loss function;

[0083] When the loss value is lower than a preset loss value, the trained convolutional neural network is used as a food image analysis model.

[0084] This embodiment collects an image dataset containing historical food images of various types of dishes. This image dataset is the basis for training a convolutional neural network and may contain historical food images of various types of dishes and corresponding labels, i.e., the optimal heat preservation temperature of each type of dish, which may be set through historical experience or manually.

[0085] Next, the collected image dataset is preprocessed, such as adjusting the size and normalization of each historical food image in the image dataset, to form a standard image dataset to meet the needs of model training. Then a convolutional neural network is constructed, which may include multiple convolutional layers, pooling layers, fully connected layers, etc., and the loss function and optimizer of the convolutional neural network are defined. The convolutional neural network is trained using the preprocessed standard image dataset, and the model is iteratively trained through forward propagation and back propagation algorithms until the model performance reaches a satisfactory level. During the training process, the model learns how to identify different types of dishes from images and determine the optimal insulation temperature corresponding to the dish type.

[0086] In this embodiment, during the training process of the convolutional neural network, a preset loss function (such as a cross entropy loss function) can be used to calculate the loss value of the model to evaluate the performance of the model. When the loss value of the model is lower than the preset loss value, it means that the model has been trained well enough and can be used as a food image analysis model in actual application scenarios.

[0087] This embodiment can improve the quality and effect of convolutional neural network training by constructing an image dataset containing historical food images of multiple types of dishes and the optimal insulation temperature of each type of dish, and preprocessing to obtain a standard image dataset, thereby improving the accuracy of food image recognition; at the same time, preprocessing and data enhancement of the image dataset, such as size unification, color space conversion, rotation, flipping, scaling and cropping, can effectively improve the model's ability to recognize different food image perspectives and details, and enhance the generalization ability of the model. Secondly, the loss function and evaluation indicators suitable for multi-category classification, such as cross entropy loss and accuracy, are used to ensure the effectiveness of the classification task. At the same time, by calculating the loss value and comparing it with the preset loss value, the progress and effect of the model training can be effectively monitored. In addition, the constructed model can not only recognize the dishes in the training dataset, but also generalize to new types of dishes that are not included in the training dataset, providing users with reference to dish ingredients and cooking method information. In short, this technical solution realizes efficient recognition and analysis of food images through deep learning and convolutional neural network technology, improves the accuracy of recognition and the generalization ability of the model, and optimizes the model training and evaluation process.

[0088] In one embodiment, the preprocessing of the image dataset to obtain a standard image dataset includes:

[0089] Performing format unification, size adjustment, color space conversion, standardization, and data enhancement processing on each historical food image in the image dataset to obtain a plurality of standard food images;

[0090] The plurality of standard food images are combined into a standard image dataset.

[0091] In this embodiment, the format unification is to ensure that the formats of all image files are consistent, which is convenient for subsequent processing. For example, historical food images in different formats (such as JPEG, PNG, BMP, etc.) can be converted into a unified format, such as all converted into JPEG format.

[0092] Resizing is to unify the size of the image to fit the input requirements of the model, that is, to resize all historical food images to the same size. For example, resize all historical food images to 224x224 pixels.

[0093] Color space conversion is to convert images from one color space to another to suit the needs of model training. For example, convert each historical food image in RGB color space to grayscale, or convert RGB to HSV color space to better capture color information.

[0094] Normalization is the process of removing scale differences in image data to make model training more stable. For example, the pixel values ​​of each historical food image can be scaled from [0, 255] to [0, 1], or more complex normalization methods can be used, such as subtracting the mean and dividing by the standard deviation.

[0095] Data augmentation is the process of increasing the diversity of the dataset and improving the generalization ability of the model by generating variant images. Data augmentation methods include operations such as rotation, flipping, scaling, cropping, adding noise, etc. to generate new versions of the image.

[0096] Finally, multiple standard food images that have undergone the above series of preprocessing methods are integrated into a new dataset to form a standard image dataset for model training. All processed images are stored in a folder or organized in the form of a database to ensure data accessibility and consistency.

[0097] In one embodiment, after calculating the loss value of the trained convolutional neural network based on a preset loss function, the method further includes:

[0098] When the loss value is not less than a preset loss value, based on the loss value, calculating a first gradient of a preset loss function with respect to a predicted output of the convolutional neural network, and a second gradient of an output layer activation function with respect to an input of the convolutional neural network;

[0099] After multiplying the first gradient by the second gradient, an output gradient of a preset loss function with respect to an output layer of the convolutional neural network is calculated;

[0100] Calculating the gradient of the output layer of the convolutional neural network with respect to the output of the hidden layer to obtain a third gradient;

[0101] Calculating the gradient of the hidden layer activation function of the convolutional neural network with respect to the input to obtain a fourth gradient;

[0102] After multiplying the output gradient, the third gradient and the fourth gradient, a target gradient is obtained;

[0103] After multiplying each model parameter of the convolutional neural network by the target gradient respectively, a plurality of target model parameters of the convolutional neural network are obtained;

[0104] Correspondingly updating the model parameters of the convolutional neural network based on the multiple target model parameters;

[0105] The convolutional neural network after updating the model parameters is trained again using the standard image data set until the loss value is lower than the preset loss value, thereby obtaining a trained food image analysis model.

[0106] This embodiment can first determine the difference between the model predicted output and the actual output, and how this difference affects the model input. When the loss value is not less than the preset loss value, the back propagation algorithm is used to calculate the first gradient of the loss function with respect to the convolutional neural network predicted output, and the second gradient of the output layer activation function with respect to the network input.

[0107] The first gradient and the second gradient are combined to calculate the gradient of the loss function with respect to the output layer of the convolutional neural network. For example, the first gradient can be multiplied by the second gradient to obtain the output gradient, which represents the sensitivity of the loss function to the output layer of the convolutional neural network.

[0108] Then, the gradient of the output layer of the convolutional neural network with respect to the output of the hidden layer is calculated, and the gradient of the activation function of the hidden layer with respect to the input is calculated. For example, the third gradient can be obtained by calculating how the loss of the output layer affects the output of the hidden layer through the back propagation algorithm. For the activation function of the hidden layer, its gradient is calculated to obtain the fourth gradient, which represents the sensitivity of the activation function to the input.

[0109] The output gradient, the third gradient, and the fourth gradient are combined to obtain the target gradient. For example, the output gradient, the third gradient, and the fourth gradient can be multiplied to obtain the target gradient, which will be used to update the model parameters of the convolutional neural network.

[0110] Preferably, after multiplying the output gradient, the third gradient and the fourth gradient to obtain the target gradient, the output gradient, the third gradient and the fourth gradient may be multiplied to obtain an initial gradient, and the initial gradient may be clipped by a value-based gradient clipping method to obtain a target gradient, including the following formula:

[0111]

[0112] Among them, the is the target gradient, g is the initial gradient, and c is a preset positive number, indicating the maximum absolute value of gradient clipping.

[0113] This value-based gradient clipping method can directly limit the size of the gradient value, ensuring that the gradient value of each parameter does not exceed a specified range. If the gradient value exceeds this range, it will be set to the boundary value of this range, thereby preventing the gradient from exploding.

[0114] When updating the model parameters according to the target gradient, each model parameter of the convolutional neural network can be multiplied by the target gradient respectively to obtain the target model parameters corresponding to each model parameter, and the model parameters of the convolutional neural network are updated according to the target model parameters. For example, an optimization algorithm (such as SGD, Adam, etc.) can be used to update the weights and biases of the convolutional neural network according to the target model parameters.

[0115] Continue to train the convolutional neural network after updating the model parameters until the loss value of the convolutional neural network is lower than the preset loss value, and obtain a trained food image analysis model. For example, the updated convolutional neural network can be retrained using a standard image dataset, and iterate continuously until the model performance reaches a satisfactory level.

[0116] Specifically, taking a two-layer convolutional neural network as an example, the forward propagation can be expressed as:

[0117] z1=W1x+b1;

[0118] a1=f(z1);

[0119] z2=W2a1+b2;

[0120] a2=f(z2);

[0121] Among them, W1, W2 are the weights of the convolutional neural network, b1, b2 are the biases of the convolutional neural network, f is the activation function of the convolutional neural network, x is the input of the convolutional neural network, and a2 is the output of the convolutional neural network.

[0122] First, the loss function L is used to calculate the difference between the predicted output a2 and the true label y:

[0123] L = L(a2,y);

[0124] Then, calculate the gradient of the loss function with respect to the output of the output layer of the convolutional neural network:

[0125]

[0126] in, is the output gradient of the loss function with respect to the output layer of the convolutional neural network, is the first gradient of the loss function with respect to the predicted output of the convolutional neural network, It is the second gradient of the output layer activation function with respect to the input of the convolutional neural network.

[0127] Next, the target gradient is calculated based on the output gradient, the third gradient, and the fourth gradient:

[0128]

[0129] in, is the target gradient, is the output gradient of the loss function with respect to the output layer of the convolutional neural network, is the third gradient of the output layer of the convolutional neural network with respect to the output of the hidden layer, It is the fourth gradient of the hidden layer activation function of the convolutional neural network with respect to the input.

[0130] Finally, each model parameter of the convolutional neural network is multiplied by the target gradient to obtain multiple target model parameters of the convolutional neural network, and the model parameters of the convolutional neural network are updated accordingly based on the multiple target model parameters.

[0131] This embodiment can improve the accuracy of the model in identifying food images through fine gradient calculation and parameter update. The trained food image analysis model can adapt to different images and environments and has good generalization ability. At the same time, iterative training and parameter update help optimize the model performance and make it converge to the optimal state faster. In addition, by accurately controlling the parameter update and training process, computing resources and time can be saved.

[0132] In one embodiment, reference Figure 2 As shown, after setting the insulation temperature of the insulation dining furnace to the optimal insulation temperature, the method further includes:

[0133] S15, setting the stirring time of the food;

[0134] S16. When the stirring time is reached, controlling a preset stirring module to stir the food on the plate.

[0135] This embodiment can determine the optimal time for stirring the food to ensure that the food can be heated evenly during the heating process, while also ensuring that the food is evenly mixed and maintains an ideal taste and temperature.

[0136] The stirring module is started at a precise stirring time to ensure that the food is stirred according to the preset time to avoid over-stirring or under-stirring. For example, the system can use a timer or program to control the stirring module (such as an electric stirring rod) to start working at a preset time to stir the food in the plate. The stirring module can be an automatic control module integrated in the heat preservation dining stove.

[0137] For example, the system can set the stirring time of braised pork to 2 minutes to ensure that the food is heated evenly during the heating process. After the braised pork is prepared and placed on the plate of the heat preservation dining stove, the smart timer of the heat preservation dining stove is started. When the smart timer reaches 2 minutes, the stirring module starts working to stir the braised pork evenly. During the stirring process, the system can also monitor the stirring speed and strength to ensure that the food is not over-stirred, thereby destroying the texture of the food. Among them, the stirring module can be an automatic stirring arm integrated in the heat preservation dining stove, or an electric stirring rod synchronized with the smart timer.

[0138] The automated stirring process of this embodiment reduces manual operations, saves time and labor, and the preset stirring time ensures the consistency of each stirring, improves the quality and taste of the food, and ensures that the food is heated evenly.

[0139] In one embodiment, reference Figure 3 As shown, after setting the insulation temperature of the insulation dining furnace to the optimal insulation temperature, the method further includes:

[0140] S17, controlling a preset camera module in real time to shoot the food on the dinner plate to obtain a third food image;

[0141] S18, receiving the third food image sent by the camera module;

[0142] S19, calling the food image analysis model to analyze the third food image and generate an analysis result;

[0143] S20. When it is determined according to the analysis result that there is no food on the dinner plate, a reminder to add food is sent to the terminal of the staff member, and the heat preservation function of the heat preservation dinner stove is turned off.

[0144] In this embodiment, the camera module can capture the state of food on the plate in real time for subsequent image analysis. For example, the camera connected to the system is started to shoot the food on the plate to obtain the image of the current food, and ensure that each shot can obtain a clear image with a suitable angle.

[0145] Then, the food image captured by the camera is transmitted to the system, and the system receives the third food image sent by the camera module for further processing. The image data can be transmitted to the system via a wireless or wired network, or directly stored in a local storage device for subsequent analysis.

[0146] Next, the third food image is analyzed using a deep learning model to identify information such as the type and quantity of the food. For example, the received third food image is input into a pre-trained food image analysis model, and the model analyzes the third food image according to the image features and outputs analysis results, such as the type of food, remaining amount, etc.

[0147] When the food on the plate is determined to be consumed based on the analysis results, a reminder to add food is sent to the staff's terminal to remind the staff to replenish the food in time, and the keep warm function is stopped to save energy. For example, when the analysis results show that there is no food on the plate, the system will automatically send a reminder message to the staff's mobile device or workstation. At the same time, the system will send a command to the heat preservation stove to turn off its keep warm function.

[0148] For example, suppose a cafeteria uses this technical solution to manage the supply of food on the plate. Whenever the food on the plate is taken by the customer, the camera installed above the plate will automatically take a picture of the current state of the food. The camera sends the captured image to the system, and the food image analysis model on the system analyzes the image to identify the type and amount of food remaining on the plate. If the analysis results show that there is no food on the plate, the system will send a reminder to the restaurant staff's mobile phone or workstation, prompting them to refill the food. At the same time, the system will automatically turn off the insulation function under the plate to save energy.

[0149] The automated food monitoring and replenishment reminders of this embodiment reduce the inspection work of staff and improve work efficiency. At the same time, by real-time monitoring of food consumption, excessive food preparation can be avoided, thereby reducing waste. In addition, turning off the keep warm function when there is no food on the plate can save energy consumption and reduce operating costs.

[0150] In one embodiment, after setting the insulation temperature of the insulation dining furnace to the optimal insulation temperature, the method further comprises:

[0151] Acquiring temperature data of food inside the heat preservation dining stove;

[0152] Extracting characteristic information of the temperature data to obtain temperature characteristics, wherein the temperature characteristics include temperature deviation, deviation change rate and integrated absolute error;

[0153] Input the temperature characteristics and the optimal insulation temperature into a pre-trained neural network model, and output the optimal PID parameters;

[0154] The parameters of the PID controller of the heat-insulating dining stove are adjusted based on the optimal PID parameters, and the adjustment parameters and system response of the PID controller of the heat-insulating dining stove are recorded.

[0155] In this embodiment, the system monitors the current temperature of the heat preservation dining oven in real time so as to control and adjust the temperature. For example, the temperature data of the food in the heat preservation dining oven can be obtained in real time through a temperature sensor or other data acquisition equipment, and the temperature data can be sent to the system so that the system can respond.

[0156] Then, key features are extracted from the raw temperature data for subsequent analysis and control. For example, the temperature data is analyzed to extract features such as temperature deviation (the difference between the actual temperature and the set temperature), deviation change rate (the rate at which the temperature deviation changes over time), and integral absolute error (accumulated temperature deviation).

[0157] The neural network model is used to calculate the optimal PID control parameters based on the current temperature characteristics and the optimal insulation temperature. The extracted temperature characteristics and the preset optimal insulation temperature are used as inputs and sent to the pre-trained neural network model to obtain the adjusted PID controller parameters to achieve more accurate temperature control. For example, the neural network model can output the adjusted optimal PID parameters, including proportional (Kp), integral (Ki) and differential (Kd) parameters, based on the input temperature characteristics and the optimal insulation temperature.

[0158] According to the optimal PID parameters provided by the neural network model, the PID controller of the heat preservation dining stove is adjusted to optimize the temperature control performance. For example, the optimal PID parameters output by the neural network model can be applied to the PID controller of the heat preservation dining stove to replace the original parameters or as a fine-tuning.

[0159] Record the adjusted PID parameters and the temperature response of the system to facilitate subsequent analysis and optimization. That is, after adjusting the PID parameters, monitor and record the temperature changes of the thermal insulation dining furnace and the response of the PID controller to monitor the temperature response status of the thermal insulation dining furnace in real time.

[0160] For example, suppose a heat preservation stove needs to be maintained at the optimal heat preservation temperature, such as 60°C. The system monitors the actual temperature of the heat preservation stove in real time through the temperature sensor, which is 58°C. The extracted temperature features include: temperature deviation (58°C-60°C=-2°C), deviation change rate (the previous deviation was -1°C, and the change rate was -1°C), and integral absolute error (the absolute value of the accumulated temperature deviation). These temperature features and the optimal heat preservation temperature of 60°C are input into the neural network model, and the model outputs the adjusted PID parameters, such as Kp=1.2, Ki=0.05, and Kd=0.1. According to the PID parameters output by the model, the PID controller of the heat preservation stove is adjusted to control the temperature more accurately. After adjustment, the new PID parameters and the temperature changes of the heat preservation stove are recorded, such as the temperature gradually stabilizes at 60°C.

[0161] This embodiment can analyze the temperature characteristics and optimal insulation temperature under different conditions through a neural network model, accurately determine the optimal PID parameters of the PID controller, and adjust the parameters of the PID controller to the optimal PID parameters in real time, so as to more accurately maintain the temperature of the insulation dining stove and reduce temperature fluctuations. In addition, the optimized PID control can reduce energy waste and improve energy utilization efficiency. The entire temperature control process does not require manual intervention, realizing intelligent and automated temperature control.

[0162] In one embodiment, calling a pre-trained food image analysis model to analyze the second food image, identifying a target dish type to which the food in the second food image belongs, and determining an optimal insulation temperature for the target dish type includes:

[0163] Calling a pre-trained food image analysis model to extract multiple key features of the second food image;

[0164] identifying multiple types of ingredients and target cooking methods of the second food image based on multiple key features of the second food image;

[0165] Determining a target dish type to which the food in the second food image belongs based on the multiple types of ingredients and the target cooking methods;

[0166] When it is determined that the target dish type is a new dish type, calculating the pixel ratio of each type of food in the second food image to all types of food, and obtaining the weight of each type of food;

[0167] Querying the optimal keeping-warm temperature corresponding to each type of food under the target cooking method;

[0168] The optimal insulation temperature of each type of food is weighted and summed with the corresponding weight to obtain the optimal insulation temperature of the target dish type.

[0169] In this embodiment, multiple key features are identified from the food image, and the key features are helpful for the subsequent identification of the type of ingredients and cooking methods. For example, a deep learning model (such as a convolutional neural network) can be used to analyze the second food image to extract key features in the second food image, and the key features may include the shape, color, texture, etc. of the food.

[0170] Then, by analyzing the extracted key features, the model can identify the types of ingredients (such as meat, vegetables, or specific food names) and cooking methods (such as stir-frying, boiling, and roasting) in the image. Combining the identified types of ingredients and cooking methods, the model can determine the type of food dish, such as Kung Pao Chicken, Braised Pork, etc.

[0171] When the target dish type is determined to be a new dish type, the number of pixels of each ingredient type and the total number of pixels of all ingredient types are calculated. Based on the number of pixels of each ingredient type and the total number of pixels of all ingredient types, the pixel proportion of each ingredient type in all ingredient types is calculated as the weight of each ingredient type, and a weight is assigned to each ingredient type in the newly identified dish type. The weight reflects the proportion of each ingredient type in the second food image and will be used for subsequent insulation temperature calculations.

[0172] Finally, the system calculates the optimal insulation temperature for each type of food under a specific cooking method. For example, the optimal insulation temperature of each type of food is weighted and summed with the corresponding weight to obtain the optimal insulation temperature of the target dish type. That is, the optimal insulation temperature of each type of food is multiplied by its weight, and then these weighted temperatures are added together to obtain the optimal insulation temperature of the entire target dish type. This temperature comprehensively considers the insulation requirements of all types of food and can accurately determine the optimal insulation temperature of a new dish type.

[0173] This embodiment can obtain the optimal insulation temperature of the entire target dish type by weighted summing the optimal insulation temperature of each type of food and the corresponding weight, so as to comprehensively consider the insulation requirements of each type of food, and can more accurately control the temperature of the entire dish, so that customers can enjoy a more stable and delicious dining experience, and the temperature of the food is always in the best state.

[0174] In one embodiment, after determining the target dish type of the food in the second food image based on the multiple types of ingredients and the target cooking methods, the method further includes:

[0175] When it is determined that the target dish type is not a new dish type, the optimal insulation temperature of the target dish type is obtained from a database.

[0176] In this embodiment, for the dish type that has been recorded in the database, the system will quickly retrieve its optimal insulation temperature so that it can be directly applied to the temperature control of the insulation dining stove. For example, based on the dish recognition result, the system will check whether the dish type already exists in the database. If it is a known dish type, the system will search the database for the optimal insulation temperature corresponding to the identified dish type. The optimal insulation temperature can be based on historical data and experience, which can ensure that the food remains in its best state during the insulation process.

[0177] For example, when a staff member puts a dish of "Kung Pao Chicken" into the heat preservation oven, the system recognizes it as "Kung Pao Chicken" through the food image analysis model. The system checks the database and finds that "Kung Pao Chicken" already exists in the database, so it is not a new dish type. The system finds out in the database that the optimal heat preservation temperature for "Kung Pao Chicken" is 60℃, and automatically sets the temperature of the heat preservation oven to 60℃.

[0178] In this embodiment, the system can quickly retrieve the optimal insulation temperature for a known dish type, reducing manual intervention and setting time, and improving overall operation efficiency. At the same time, by accurately controlling the insulation temperature, the best taste and nutrition of the food can be maintained, improving the customer's dining experience. In addition, the system can automatically manage the insulation temperature of different dishes, reducing the need for manual monitoring and realizing intelligent management.

[0179] Please refer to Figure 4 The embodiment of the present invention further provides an artificial intelligence-based heat preservation dining stove operation control system, comprising:

[0180] The control module 41 is used for controlling a preset camera module to photograph the food on the dinner plate to obtain a first food image after the food is placed on the dinner plate of the heat preservation dining oven;

[0181] A preprocessing module 42, configured to receive the first food image sent by the camera module, and preprocess the first food image to obtain a second food image;

[0182] An analysis module 43 is used to call a pre-trained food image analysis model to analyze the second food image, identify the target dish type to which the food in the second food image belongs, and determine the optimal insulation temperature of the target dish type, wherein the food image analysis model is a convolutional neural network, and is used to identify the dish type in the food image and determine the optimal insulation temperature of the dish type;

[0183] The setting module 44 is used to set the insulation temperature of the insulation dining stove to the optimal insulation temperature.

[0184] The artificial intelligence-based insulation dining furnace operation control system provided by the present invention receives a first food image sent by a camera module, pre-processes the first food image to obtain a second food image, calls a pre-trained food image analysis model to analyze the second food image, identifies the target dish type to which the food in the second food image belongs and determines the optimal insulation temperature of the target dish type, sets the insulation temperature of the insulation dining furnace to the optimal insulation temperature, and uses a convolutional neural network to analyze the food image, so as to accurately identify the dish type and determine the optimal insulation temperature of the dish type, and automatically sets the insulation temperature of the insulation dining furnace to the optimal insulation temperature, thereby realizing automatic and intelligent control of the temperature of the insulation dining furnace by artificial intelligence, and the whole process does not require manual intervention, thereby improving operational convenience and efficiency.

[0185] Regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0186] In one embodiment, the present invention further proposes a storage medium storing computer-readable instructions, which, when executed by one or more processors, enables one or more processors to execute the above-mentioned artificial intelligence-based heat preservation dining stove operation control method. The storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0187] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0188] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0189] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for controlling the operation of a heat preservation dining stove based on artificial intelligence, characterized in that: include: When food is placed on a dinner plate of the heat preservation dining oven, a preset camera module is controlled to shoot the food on the dinner plate to obtain a first food image; receiving a first food image sent by the camera module, and preprocessing the first food image to obtain a second food image; Calling a pre-trained food image analysis model to analyze the second food image, identifying a target dish type to which the food in the second food image belongs and determining an optimal heat preservation temperature for the target dish type, wherein the food image analysis model is a convolutional neural network, and is used to identify the dish type in the food image and determine the optimal heat preservation temperature for the dish type; The insulation temperature of the insulation dining stove is set to the optimal insulation temperature.

2. The artificial intelligence-based heat preservation dining stove operation control method according to claim 1 is characterized in that: Before calling the pre-trained food image analysis model to analyze the second food image, identifying the target dish type to which the food in the second food image belongs, and determining the optimal insulation temperature of the target dish type, the method further includes: Collecting image datasets of different dish types, wherein the image datasets include historical food images of multiple dish types and the optimal keeping temperature of each dish type; Preprocessing the image data set to obtain a standard image data set; Using the standard image data set to train a convolutional neural network, and calculating a loss value of the trained convolutional neural network based on a preset loss function; When the loss value is lower than a preset loss value, the trained convolutional neural network is used as a food image analysis model.

3. The artificial intelligence-based heat preservation dining stove operation control method according to claim 2 is characterized in that: The preprocessing of the image data set to obtain a standard image data set includes: Performing format unification, size adjustment, color space conversion, standardization, and data enhancement processing on each historical food image in the image dataset to obtain a plurality of standard food images; The plurality of standard food images are combined into a standard image dataset.

4. The artificial intelligence-based heat preservation dining stove operation control method according to claim 2 is characterized in that: After calculating the loss value of the trained convolutional neural network based on a preset loss function, the method further includes: When the loss value is not less than a preset loss value, based on the loss value, calculating a first gradient of a preset loss function with respect to a predicted output of the convolutional neural network, and a second gradient of an output layer activation function with respect to an input of the convolutional neural network; After multiplying the first gradient by the second gradient, an output gradient of a preset loss function with respect to an output layer of the convolutional neural network is calculated; Calculating the gradient of the output layer of the convolutional neural network with respect to the output of the hidden layer to obtain a third gradient; Calculating the gradient of the hidden layer activation function of the convolutional neural network with respect to the input to obtain a fourth gradient; After multiplying the output gradient, the third gradient and the fourth gradient, a target gradient is obtained; After multiplying each model parameter of the convolutional neural network by the target gradient respectively, a plurality of target model parameters of the convolutional neural network are obtained; Correspondingly updating the model parameters of the convolutional neural network based on the multiple target model parameters; The convolutional neural network after updating the model parameters is trained again using the standard image data set until the loss value is lower than the preset loss value, thereby obtaining a trained food image analysis model.

5. The artificial intelligence-based heat preservation dining stove operation control method according to claim 1 is characterized in that: After the heat preservation temperature of the heat preservation dining furnace is set to the optimal heat preservation temperature, the method further comprises: Setting a blending time for the food; When the stirring time is reached, a preset stirring module is controlled to stir the food on the plate.

6. The artificial intelligence-based heat preservation dining stove operation control method according to claim 1 is characterized in that: After the heat preservation temperature of the heat preservation dining furnace is set to the optimal heat preservation temperature, the method further comprises: controlling a preset camera module in real time to shoot the food on the dinner plate to obtain a third food image; receiving the third food image sent by the camera module; calling the food image analysis model to analyze the third food image and generate an analysis result; When it is determined according to the analysis result that there is no food on the dinner plate, a reminder to add food is sent to the terminal to which the staff belongs, and the heat preservation function of the heat preservation dinner stove is turned off.

7. The artificial intelligence-based heat preservation dining stove operation control method according to claim 1 is characterized in that: After the heat preservation temperature of the heat preservation dining furnace is set to the optimal heat preservation temperature, the method further comprises: Acquiring temperature data of food inside the heat preservation dining stove; Extracting characteristic information of the temperature data to obtain temperature characteristics, wherein the temperature characteristics include temperature deviation, deviation change rate and integrated absolute error; Input the temperature characteristics and the optimal insulation temperature into a pre-trained neural network model, and output the optimal PID parameters; The parameters of the PID controller of the heat-insulating dining stove are adjusted based on the optimal PID parameters, and the adjustment parameters and system response of the PID controller of the heat-insulating dining stove are recorded.

8. The artificial intelligence-based heat preservation dining stove operation control method according to claim 1, characterized in that: The calling of a pre-trained food image analysis model to analyze the second food image, identifying a target dish type to which the food in the second food image belongs and determining an optimal heat preservation temperature for the target dish type includes: Calling a pre-trained food image analysis model to extract multiple key features of the second food image; identifying multiple types of ingredients and target cooking methods of the second food image based on multiple key features of the second food image; Determining a target dish type to which the food in the second food image belongs based on the multiple types of ingredients and the target cooking methods; When it is determined that the target dish type is a new dish type, calculating the pixel ratio of each type of food in the second food image to all types of food, and obtaining the weight of each type of food; Querying the optimal keeping-warm temperature corresponding to each type of food under the target cooking method; The optimal insulation temperature of each type of food is weighted and summed with the corresponding weight to obtain the optimal insulation temperature of the target dish type.

9. The artificial intelligence-based heat preservation dining stove operation control method according to claim 8, characterized in that: After determining the target dish type of the food in the second food image based on the multiple types of ingredients and the target cooking methods, the method further includes: When it is determined that the target dish type is not a new dish type, the optimal insulation temperature of the target dish type is obtained from a database.

10. An artificial intelligence-based heat preservation dining stove operation control system, characterized in that: include: A control module, configured to control a preset camera module to photograph the food on the dinner plate to obtain a first food image after the food is placed on the dinner plate of the heat preservation dining oven; a preprocessing module, configured to receive the first food image sent by the camera module, and preprocess the first food image to obtain a second food image; an analysis module, configured to call a pre-trained food image analysis model to analyze the second food image, identify a target dish type to which the food in the second food image belongs, and determine an optimal heat preservation temperature for the target dish type, wherein the food image analysis model is a convolutional neural network, configured to identify the dish type in the food image and determine an optimal heat preservation temperature for the dish type; A setting module is used to set the insulation temperature of the insulation dining stove to the optimal insulation temperature.

Images