Cooking control method and device and cooking device

Abstract

The invention provides a cooking control method and device and a cooking device.The cooking control method comprises the steps that first cooking image data of a target object at a preset historical moment and second cooking image data of the target object at the current moment are obtained, and a preset time interval exists between the preset historical moment and the current moment; based on the first cooking image data and the second cooking image data, obtaining corresponding prediction state data of the target object at a target moment, wherein the target moment is a moment having the preset time interval with the current moment after the current moment; and under the condition that the prediction state data meets a preset adjustment condition, generating cooking adjustment data based on the prediction state data. Therefore, the state of the food at the future moment in the cooking process can be predicted, the cooking process can be conveniently regulated and controlled in time in the food cooking process, the delay problem of the cooking device is solved, and the flavor and appearance of the food are guaranteed.

Description

Technical Field

[0001] The present disclosure relates to the field of cooking control, and in particular to a cooking control method, device and cooking device. Background Art

[0002] Existing methods for judging the category of ingredients and the state of the cooking process based on visual information generally use the current image information of the cooking ingredients or the image information of the past period of time as input, and determine the current type of ingredients and cooking state through detection algorithms. However, the state of the food cooking process is greatly affected by the proportion of ingredients and cooking conditions. The control of cooking conditions has a certain delay. It is difficult to ensure the flavor and appearance of the ingredients at the end of cooking by controlling the firepower based on the current state of the food cooking process. Summary of the invention

[0003] In order to solve the above technical problems, the present disclosure proposes a cooking control method, device and cooking device.

[0004] According to a first aspect of the present disclosure, there is provided a cooking control method, comprising:

[0005] Acquire first cooking image data and second cooking image data corresponding to a target object, wherein the first cooking image data is obtained by collecting images of the target object at a preset historical moment, and the second cooking image data is obtained by collecting images of the target object at a current moment, and there is a preset time interval between the preset historical moment and the current moment;

[0006] Performing state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target time, wherein the target time is a time after the current time and with a preset time interval from the current time;

[0007] In a case where the predicted state data meets a preset adjustment condition, cooking adjustment data is generated based on the predicted state data, and the cooking adjustment data is used to adjust a cooking control parameter.

[0008] Optionally, performing state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target time includes:

[0009] Performing feature extraction processing on the first cooking image data and the second cooking image data to generate first feature data and second feature data respectively, wherein the first feature data includes a plurality of first feature graphs, and the second feature data includes a plurality of second feature graphs;

[0010] Performing identification processing on the first feature data and the second feature data to generate first converted feature data and second converted feature data respectively, wherein the identification processing is used to identify the plurality of first feature graphs and the plurality of second feature graphs respectively;

[0011] State prediction is performed based on the first conversion feature data and the second conversion feature data to obtain predicted state data corresponding to the target object at the target time.

[0012] Optionally, the performing identification processing on the first feature data and the second feature data to generate first converted feature data and second converted feature data respectively includes:

[0013] Acquire first identification data corresponding to the first feature data and second identification data corresponding to the second feature data, wherein the first identification data includes first identification parameters corresponding to the plurality of first feature graphs, and the second identification data includes second identification parameters corresponding to the plurality of second feature graphs;

[0014] The first feature data and the second feature data are identified according to the first identification data and the second identification data, respectively, to generate first converted feature data and second converted feature data.

[0015] Optionally, the acquiring first identification data corresponding to the first feature data and second identification data corresponding to the second feature data includes:

[0016] acquiring a first marking constant corresponding to the first cooking image data and a second marking constant corresponding to the second cooking image data;

[0017] Obtaining first identification data according to the first marking constant and the channel positions respectively corresponding to the plurality of first feature maps;

[0018] Second identification data is obtained according to the second marking constant and the channel positions respectively corresponding to the plurality of second feature maps.

[0019] Optionally, the predicted state data is used to indicate a predicted cooking state category of the target object at a target time, and when the predicted state data satisfies a preset adjustment condition, generating cooking adjustment data based on the predicted state data includes:

[0020] When the predicted state data indicates that the target object is in a preset cooking state category at a target time and the cooking time corresponding to the target time is less than the preset cooking time, cooking adjustment data is generated based on the predicted state data.

[0021] Optionally, the predicted state data includes predicted cooking state parameters of the target object at the target time, and when the predicted state data satisfies a preset adjustment condition, generating cooking adjustment data based on the predicted state data includes:

[0022] When the degree of deviation of the predicted cooking state parameter from the preset cooking state parameter corresponding to the target time is within a preset range, cooking adjustment data is generated based on the predicted cooking state parameter.

[0023] Optionally, performing state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target time includes:

[0024] Using the state prediction model to perform state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target time;

[0025] The method further comprises:

[0026] According to the sample data, the initial prediction model is trained based on back propagation to obtain the state prediction model. The loss parameter of the back propagation is obtained according to the predicted state loss and the binary cross entropy loss between the predicted state data and the corresponding actual state data. The actual state data is used to indicate the cooking state of the target object at the moment corresponding to the predicted state data.

[0027] Optionally, the method further includes:

[0028] According to the second cooking image data, obtaining the real-time state data corresponding to the target object at the current moment;

[0029] When the predicted status data and the real-time status data meet the preset alarm conditions, alarm information is generated.

[0030] According to a second aspect of the present disclosure, there is provided a cooking control device, comprising:

[0031] An image data acquisition module, configured to acquire first cooking image data and second cooking image data corresponding to a target object, wherein the first cooking image data is obtained by acquiring an image of the target object at a preset historical moment, and the second cooking image data is obtained by acquiring an image of the target object at a current moment, and there is a preset time interval between the preset historical moment and the current moment;

[0032] a state prediction module, configured to perform state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target time, wherein the target time is a time after the current time and having a preset time interval with the current time;

[0033] The cooking adjustment module is used to generate cooking adjustment data based on the predicted state data when the predicted state data meets the preset adjustment conditions, and the cooking adjustment data is used to adjust the cooking control parameters.

[0034] According to a third aspect of the present disclosure, a cooking device is provided, comprising a cooking component and at least one processor, and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the at least one processor is used to control the cooking component to perform cooking, and the at least one processor implements the cooking control method described in the above technical solution by executing the instructions stored in the memory.

[0035] According to a fourth aspect of the present disclosure, a cooking device is provided, comprising a cooking component and a cooking control device as described in the above technical solution, wherein the cooking control device is used to control the cooking component to perform cooking.

[0036] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

[0037] The implementation of this disclosure has the following beneficial effects:

[0038] The present disclosure provides a cooking control method, device and cooking device, comprising: obtaining first cooking image data and second cooking image data corresponding to a target object, wherein the first cooking image data is obtained by performing image acquisition on the target object at a preset historical moment, and the second cooking image data is obtained by performing image acquisition on the target object at a current moment, and there is a preset time interval between the preset historical moment and the current moment; performing state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target moment, wherein the target moment is a moment after the current moment and has a preset time interval with the current moment; and generating cooking adjustment data based on the predicted state data when the predicted state data meets a preset adjustment condition. Thus, it is possible to make a prediction for the state of food at a future moment during the cooking process, so as to facilitate timely regulation of the cooking process during the cooking process of the food, overcome the delay problem of the cooking device, and ensure the flavor and appearance of the food.

[0039] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions and advantages in the embodiments of this specification or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 A schematic diagram showing a flow chart of a cooking control method according to an embodiment of the present disclosure;

[0042] Figure 2 A schematic diagram showing a process of determining predicted state data corresponding to a target object at a target time according to an embodiment of the present disclosure;

[0043] Figure 3 A schematic diagram showing a process of generating first conversion feature data and second conversion feature data according to an embodiment of the present disclosure;

[0044] Figure 4 A schematic diagram showing a process of obtaining first identification data and second identification data according to an embodiment of the present disclosure is shown;

[0045] Figure 5 A schematic diagram showing a process of generating alarm information according to an embodiment of the present disclosure;

[0046] Figure 6 A schematic structural diagram of a cooking control device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0049] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0050] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0051] The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set consisting of A, B, and C.

[0052] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0053] Figure 1A flow chart of a cooking control method according to an embodiment of the present disclosure is shown. The cooking control method of the embodiment of the present invention can be applied to cooking equipment, smart terminals or background servers, such as a control unit of a cooking device, or a smart terminal that is communicatively connected to a cooking device; a cooking device is a device for cooking food or a target object, for example, a cooking device is an oven, a steam oven, an air fryer, a microwave oven or an electric stove, and the smart terminal or background server is at least used to control the cooking device to cook and control the cooking of the cooking device. The smart terminal can be a PC, a mobile phone, a tablet computer or other smart electronic device. Exemplarily, the smart terminal or background server sends cooking adjustment data to the cooking device so that the cooking device adjusts the cooking control parameters according to the received cooking adjustment data. The background server is, for example, a cloud server. This specification provides method operation steps such as embodiments or flow charts, but more or fewer operation steps may be included based on conventional or non-creative labor. The order of steps listed in the embodiments is only one way of executing the steps among many orders, and does not represent the only order of execution. When the system or server product is executed in practice, it can be executed in the order of the methods shown in the embodiments or the drawings or in parallel (for example, in a parallel processor or multi-threaded processing environment). Figure 1 As shown, the cooking control method comprises:

[0054] Step S101, obtaining first cooking image data and second cooking image data corresponding to a target object, wherein the first cooking image data is obtained by capturing an image of the target object at a preset historical moment, and the second cooking image data is obtained by capturing an image of the target object at a current moment, and there is a preset time interval between the preset historical moment and the current moment.

[0055] Specifically, during the cooking process of the target object, the image acquisition component periodically acquires images of the target object to generate cooking image data corresponding to the target object, and the cooking image data acquired at a historical moment separated from the current moment by a preset time interval T* is used as the first cooking image data, and the cooking image data acquired at the current moment is used as the second cooking image data. The period of image acquisition of the target object is less than or equal to T*, for example, the image acquisition period is 0.1T*, 0.2T*, 0.25T*, 0.5T* or T*, thereby predicting the predicted state of the target object based on the cooking image data acquired in each image acquisition period. Preferably, the preset time interval is n times the image acquisition period, and n is an integer greater than or equal to 1.

[0056] In an alternative embodiment, during the cooking process of the target object, the image acquisition component is used to acquire images of the target object to generate cooking video data, and the cooking video data is periodically sampled to obtain cooking image data corresponding to the target object, and the cooking image data corresponding to the historical moment separated by a preset time interval T* from the current moment is used as the first cooking image data, and the cooking image data corresponding to the current moment is used as the second cooking image data. The sampling period for sampling the cooking video data is less than or equal to T*, for example, the sampling period is 0.1T*, 0.2T*, 0.25T*, 0.5T* or T*, thereby, the predicted state of the target object can be predicted based on the cooking image data obtained in each sampling period. Preferably, the preset time interval is m times the sampling period, and m is an integer greater than or equal to 1.

[0057] Optionally, the preset time interval T* is 3s-120s, for example, 5s, 10s, 12s, 15s, 20s, 30s, 45s, 60s, 80s or 100s.

[0058] Step S102, performing state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target moment, wherein the target moment is a moment after the current moment and with a preset time interval from the current moment.

[0059] Specifically, state prediction processing is performed based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at the target moment, and the predicted state data is used to indicate the cooking state of the target object at the target moment, such as the cooking state category or cooking state parameters of the target object. The target moment is a moment after the current moment, and there is a preset time interval between the target moment and the current moment. For example, the target moment is a moment that is 5s, 10s, 15s, 20s or 30s after the current moment.

[0060] Exemplarily, the cooking status categories include raw and cooked. Optionally, the cooking status categories also include at least one of half-cooked, burnt and mushy.

[0061] Exemplarily, the cooking state parameter is the maturity of the target object. The cooking state parameter can be used to indicate the maturity of the target object. For example, the value range of the cooking state parameter is 0-1, 0 indicates that the target object is in a raw state, 0.1, 0.3, 0.5, 0.8 and 1 respectively indicate that the target object is in a 10% cooked, 30% cooked, 50% cooked, 80% cooked and 100% cooked state. In other optional embodiments, the value range of the cooking state parameter is 0-1.2, 0 indicates that the target object is in a raw state, 0.1, 0.3, 0.5, 0.8 and 1 respectively indicate that the target object is in a 10% cooked, 30% cooked, 50% cooked, 80% cooked and 100% cooked state. When the cooking state parameter is greater than 1, it indicates that the target object is in an overcooked state, such as overcooked barbecue or burnt biscuits. The value range of the cooking state parameter can also be represented by other numerical values, which can be used to indicate different cooking state parameters of the target object from raw to cooked, from raw to burnt or from raw to mushy.

[0062] Step S103, generating cooking adjustment data based on the predicted state data when the predicted state data meets the preset adjustment condition, wherein the cooking adjustment data is used to adjust the cooking control parameters.

[0063] Specifically, the preset adjustment condition is used to determine whether the predicted cooking state corresponding to the target object at the target time deviates from the expected cooking state. The predicted state data satisfies the preset adjustment condition, indicating that the predicted cooking state corresponding to the target object at the target time deviates from the expected cooking state, and then the cooking adjustment data is generated based on the predicted state data, so that the cooking device or cooking component adjusts the cooking control parameters according to the cooking adjustment data, and the cooking control parameters are used to control the cooking device to cook. Exemplarily, the cooking control parameters include at least one of heating power, cooking temperature, cooking time and fan speed.

[0064] Therefore, the implementation method provided by the present disclosure determines the predicted state data corresponding to the target object at the target moment based on the first cooking image data and the second cooking image data corresponding to the target object, and generates cooking adjustment data based on the predicted state data when the predicted state data satisfies a preset adjustment condition. Therefore, cooking control parameters can be adjusted based on the predicted state data of the target object, thereby avoiding poor cooking effects caused by the control delay in the cooking environment provided by the cooking device, and improving the timeliness and efficiency of cooking control.

[0065] In an optional embodiment, if Figure 2 As shown, the state prediction processing is performed based on the first cooking image data and the second cooking image data to obtain the predicted state data corresponding to the target object at the target time, including:

[0066] Step S11 , performing feature extraction processing on the first cooking image data and the second cooking image data to generate first feature data and second feature data respectively, wherein the first feature data includes a plurality of first feature graphs, and the second feature data includes a plurality of second feature graphs.

[0067] Specifically, feature extraction processing is performed on the first cooking image data, such as convolution processing, to obtain first feature data, where the first feature data is multi-channel feature map data. Exemplarily, the size of the multi-channel feature map is H×W×C, where H, W, and C are respectively the height, width, and number of channels of the multi-channel feature map. For example, H is 320-1280, such as 640, w is 320-1280, such as 640, and C is 16-128, such as 32 or 64. The feature map of each channel in the first feature data constitutes a first feature map, and the first feature data includes C first feature maps.

[0068] The second cooking image data is subjected to the same feature extraction process as the first cooking image data to obtain second feature data. Accordingly, the second feature data is multi-channel feature map data, and the size of the second feature data is the same as that of the first feature data. The feature map of each channel in the second feature data constitutes a second feature map, and the second feature data includes C second feature maps.

[0069] Step S12, performing identification processing on the first feature data and the second feature data to generate first converted feature data and second converted feature data respectively, wherein the identification processing is used to identify the plurality of first feature graphs and the plurality of second feature graphs respectively.

[0070] Specifically, each first feature graph in the first feature data and each second feature graph in the second feature data are identified respectively, each feature graph of the multiple first feature graphs and the multiple second feature graphs corresponds to an identification parameter, and the feature graphs are identified using the corresponding identification parameters to obtain first converted feature data corresponding to the first feature data and second converted feature data corresponding to the second feature data.

[0071] Optionally, the value range of the identification parameter is -1 to 1. In other optional implementations, the upper limit value and the lower limit value of the identification parameter are determined according to the multiple first feature graphs and the multiple second feature graphs, for example, according to the distribution state or value level of the values ​​in the multiple first feature graphs and the multiple second feature graphs.

[0072] Step S13: performing state prediction based on the first conversion feature data and the second conversion feature data to obtain predicted state data corresponding to the target object at a target time.

[0073] Specifically, a state prediction is performed based on the first conversion feature data and the second conversion feature data to obtain the predicted state data corresponding to the target object at the target time. Optionally, the first conversion feature data and the second conversion feature data are sequentially combined and input as one input data into a state prediction model for state prediction, or the first conversion feature data and the second conversion feature data are input as two different input data into a state prediction model for state prediction.

[0074] In an optional embodiment, if Figure 3 As shown, the identification processing of the first feature data and the second feature data to generate first conversion feature data and second conversion feature data respectively includes:

[0075] Step S21, obtaining first identification data corresponding to the first feature data and second identification data corresponding to the second feature data, wherein the first identification data includes first identification parameters corresponding to the multiple first feature maps respectively, and the second identification data includes second identification parameters corresponding to the multiple second feature maps respectively.

[0076] Specifically, the first identification data is used to mark the plurality of first characteristic graphs in the first characteristic data, and the second identification data is used to mark the plurality of second characteristic graphs in the second characteristic data. Exemplarily, the first identification data includes C first identification parameters T en (C1, i), that is, T en (C1,1),T en (C1,2),T en (C1,3)……T en (C1, C), the second identification data includes C second identification parameters T en (C2, i), that is, T en (C2,1),T en (C2,2),T en (C2,3)……T en (C2, C), C1 is the first marking constant corresponding to the first feature data, for example, C1 is -1, 0 or other values, C2 is the second marking constant corresponding to the second feature data, for example, C2 is 0, 1 or other values, C1 and C2 have different values, i represents the i-th channel, the value range of i is 1 to C, i is an integer, C is the total number of first feature maps in the first feature data or the total number of second feature maps in the second feature data, and the total number of first feature maps in the first feature data is the same as the total number of second feature maps in the second feature data.

[0077] Step S22: Identify the first characteristic data and the second characteristic data according to the first identification data and the second identification data, respectively, to generate first converted characteristic data and second converted characteristic data.

[0078] Specifically, each first characteristic graph in the first characteristic data is respectively identified according to the first identification data to obtain first converted characteristic data. Specifically, the first identification parameter T is used. en (C1, i) identifies the i-th first feature graph in the first feature data, thereby identifying each first feature graph in the first feature data respectively.

[0079] Each second characteristic graph in the second characteristic data is identified according to the second identification data to obtain second converted characteristic data. Specifically, the second identification parameter T en (C2, i) identifies the i-th second feature graph in the second feature data, thereby identifying each second feature graph in the second feature data respectively.

[0080] In an optional embodiment, if Figure 4 As shown, the obtaining of the first identification data corresponding to the first feature data and the second identification data corresponding to the second feature data, or before the obtaining of the first identification data corresponding to the first feature data and the second identification data corresponding to the second feature data, includes:

[0081] Step S31, acquiring a first marking constant corresponding to the first cooking image data and a second marking constant corresponding to the second cooking image data.

[0082] Specifically, different marking constants are assigned to the first cooking image data and the second cooking image data. For example, based on the different times corresponding to the first cooking image data and the second cooking image data, the first marking constant is set to 0 and the second marking constant is set to 1, or the first marking constant is set to -1 and the second marking constant is set to 1.

[0083] Step S32: Obtain first identification data according to the first marking constant and the channel positions respectively corresponding to the plurality of first feature maps.

[0084] Specifically, the first marking constant and the channel position corresponding to each first feature map are converted as variables to obtain first identification data.

[0085] Exemplarily, the calculation formula of the first identification data is:

[0086]

[0087] Among them, T en(C1, i) is the i-th first identification parameter in the first identification data, that is, the first identification parameter corresponding to the i-th first feature map.

[0088] Step S33, obtaining second identification data according to the second marking constant and the channel positions respectively corresponding to the plurality of second feature maps.

[0089] Specifically, the second marking constant and the channel position corresponding to each second feature map are converted as variables to obtain second identification data.

[0090] Exemplarily, the calculation formula of the second identification data is:

[0091]

[0092] Among them, T en (C1, i) is the i-th second identification parameter in the second identification data, that is, the second identification parameter corresponding to the i-th second feature map.

[0093] In alternative implementations, other functional relationships may be used to determine the first identification data and the second identification data, and the above description is not intended to be the sole limitation on the implementations of the present disclosure.

[0094] In an alternative implementation, the first identification data and the second identification data are stored in a memory, and the pre-stored first identification data and the second identification data are acquired by performing a read operation on the memory.

[0095] In an optional embodiment, the predicted state data is used to indicate a predicted cooking state category of the target object at a target time, and when the predicted state data satisfies a preset adjustment condition, generating cooking adjustment data based on the predicted state data includes:

[0096] When the predicted state data indicates that the target object is in a preset cooking state category at a target time and the cooking time corresponding to the target time is less than the preset cooking time, cooking adjustment data is generated based on the predicted state data.

[0097] Specifically, a preset cooking state category is set corresponding to a preset cooking time of the target object, and it is determined whether the cooking control parameters need to be adjusted according to the predicted state data of the target object at the target time.

[0098] Exemplarily, the preset cooking time is 50 minutes, and the preset cooking status category of the target object after 50 minutes is cooked. If the predicted status data indicates that the cooking status category of the target object at the target moment is cooked, and the cooking time corresponding to the target moment is less than the preset cooking time of 50 minutes, for example, the cooking time corresponding to the target moment is 45 minutes or 48 minutes, then the cooking speed of the target object is too fast, and the cooking power needs to be reduced or the cooking temperature needs to be lowered.

[0099] Optionally, the entire cooking process is divided into multiple cooking time periods, each cooking time period corresponding to a different preset cooking state of the target object. When the preset cooking state category of the target object at the target moment indicated by the predicted state data does not match the cooking time corresponding to the target moment, cooking adjustment data is generated based on the predicted state data to reduce the cooking power and lower the cooking temperature, or increase the cooking power and increase the cooking temperature.

[0100] In an optional embodiment, the predicted state data includes predicted cooking state parameters of the target object at the target time, and when the predicted state data satisfies a preset adjustment condition, generating cooking adjustment data based on the predicted state data includes:

[0101] When the degree of deviation of the predicted cooking state parameter from the preset cooking state parameter corresponding to the target time is within a preset range, cooking adjustment data is generated based on the predicted cooking state parameter.

[0102] Specifically, the predicted cooking state parameter is a quantitative parameter used to indicate the degree of maturity of the target object. For example, the predicted cooking state parameter is 0.1, 0.3, 0.5 or 0.8, indicating that the target object is 10% cooked, 30% cooked, 50% cooked or 80% cooked, respectively.

[0103] The corresponding relationship between the preset cooking state parameter and the cooking time is preset, and the deviation degree can be the predicted cooking state parameter p 1 The preset cooking state parameter p corresponding to the target time 0 The difference between 1 -p 0 or degree of deviation (p 1 -p 0 ) / p 0 For example, in the difference p 1 -p 0 is greater than the first preset threshold or less than the second preset threshold, or the degree of deviation (p 1 -p 0 ) / p 0When the difference p is greater than the third preset threshold or less than the fourth preset threshold, cooking adjustment data is generated based on the predicted cooking state parameter, such as reducing the cooking power, lowering the cooking temperature, or increasing the cooking power, and raising the cooking temperature. 1 -p 0 or the degree of deviation (p 1 -p 0 ) / p 0 Generate cooking adjustment data to improve the accuracy of cooking adjustments.

[0104] In an optional embodiment, the performing state prediction processing based on the first cooking image data and the second cooking image data to obtain the predicted state data corresponding to the target object at the target time includes:

[0105] The state prediction model is used to perform state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at the target time.

[0106] Accordingly, the method further includes:

[0107] According to the sample data, the initial prediction model is trained based on back propagation to obtain the state prediction model. The loss parameter of the back propagation is obtained according to the predicted state loss and the binary cross entropy loss between the predicted state data and the corresponding actual state data. The actual state data is used to indicate the cooking state of the target object at the moment corresponding to the predicted state data.

[0108] Specifically, the sample data includes multiple cooking samples, each cooking sample includes first cooking image data of the cooking object at the first moment, second cooking image data at the second moment, and third cooking image data / cooking status data at the third moment, the cooking status data represents the actual cooking status of the cooking object at the third moment, and may be a cooking status category or cooking status parameter of the cooking object at the third moment, the first moment, the second moment, and the third moment are three moments arranged in chronological order, and the time interval between two adjacent moments is a preset time interval.

[0109] The binary cross entropy loss BCELoss is calculated according to the following formula:

[0110] BCELoss=γ t′ ·log(σ t′ )+(1-γ t′ )·log(1-σ t′ )

[0111] Among them, γ t′is the cooking state data of the cooking object at the third moment or the actual state data of the cooking object obtained according to the third cooking image data, σ t′ The third moment predicted state data corresponding to the cooking object.

[0112] Optionally, the loss parameter L is calculated according to the following formula:

[0113] L=L0+k·log BCELoss+q

[0114] Wherein, L0 is the predicted state loss, k is a constant, ranging from 1 to 5, such as 2 or 3, and q is an adjustment constant, ranging from -1 to 1, such as 0, 0.2 or 0.5.

[0115] In an optional embodiment, if Figure 5 As shown, the method also includes:

[0116] Step S41: obtaining the real-time status data corresponding to the target object at the current moment according to the second cooking image data.

[0117] Specifically, the real-time status data represents the real-time cooking status of the target object at the current moment, and the real-time status data may be the real-time cooking status category or real-time cooking status parameter of the target object at the current moment.

[0118] Step S42, generating alarm information when the predicted status data and the real-time status data meet preset alarm conditions.

[0119] Specifically, when the predicted status data and the real-time status data meet the preset alarm conditions, alarm information is generated so that the alarm component issues an alarm according to the alarm information.

[0120] Exemplarily, when the real-time status data is in focus and the predicted status data is in focus or blurred, an alarm message is generated; or, when the real-time status data is 100% and the predicted status data is 110%, an alarm message is generated.

[0121] Figure 6 A block diagram of a cooking control device according to an embodiment of the present disclosure is shown; Figure 6 As shown, the above device comprises:

[0122] An image data acquisition module, configured to acquire first cooking image data and second cooking image data corresponding to a target object, wherein the first cooking image data is obtained by acquiring an image of the target object at a preset historical moment, and the second cooking image data is obtained by acquiring an image of the target object at a current moment, and there is a preset time interval between the preset historical moment and the current moment;

[0123] a state prediction module, configured to perform state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target time, wherein the target time is a time after the current time and having a preset time interval with the current time;

[0124] The cooking adjustment module is used to generate cooking adjustment data based on the predicted state data when the predicted state data meets the preset adjustment conditions, and the cooking adjustment data is used to adjust the cooking control parameters.

[0125] In some embodiments, the functions or modules / units included in the cooking control device provided in the embodiments of the present disclosure can be used to execute the cooking control method described in the above embodiments. Its specific implementation can refer to the description of the above embodiments, and for the sake of brevity, it will not be repeated here.

[0126] An embodiment of the present disclosure also proposes a cooking device, comprising a cooking component and at least one processor, and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the at least one processor is used to control the cooking component to perform cooking, and the at least one processor implements the cooking control method described in the above technical solution by executing the instructions stored in the memory.

[0127] The disclosed embodiment also provides a cooking device, comprising the cooking control device as described in the above technical solution. The cooking device also includes a cooking component, and the cooking control device can control the cooking component to cook.

[0128] The embodiment of the present disclosure also proposes an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor implements the cooking control method by executing the instructions stored in the memory.

[0129] The electronic device may be provided as a terminal, a server, or a device in other forms.

[0130] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A cooking control method, characterized in that: include: Acquire first cooking image data and second cooking image data corresponding to a target object, wherein the first cooking image data is obtained by collecting images of the target object at a preset historical moment, and the second cooking image data is obtained by collecting images of the target object at a current moment, and there is a preset time interval between the preset historical moment and the current moment; Performing state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target time, wherein the target time is a time after the current time and with a preset time interval from the current time; In a case where the predicted state data meets a preset adjustment condition, cooking adjustment data is generated based on the predicted state data, and the cooking adjustment data is used to adjust a cooking control parameter.

2. The cooking control method according to claim 1, characterized in that: The performing state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at the target time includes: Performing feature extraction processing on the first cooking image data and the second cooking image data to generate first feature data and second feature data respectively, wherein the first feature data includes a plurality of first feature graphs, and the second feature data includes a plurality of second feature graphs; Performing identification processing on the first feature data and the second feature data to generate first converted feature data and second converted feature data respectively, wherein the identification processing is used to identify the plurality of first feature graphs and the plurality of second feature graphs respectively; State prediction is performed based on the first conversion feature data and the second conversion feature data to obtain predicted state data corresponding to the target object at the target time.

3. The cooking control method according to claim 2, characterized in that: The step of performing identification processing on the first feature data and the second feature data to generate first converted feature data and second converted feature data respectively includes: Acquire first identification data corresponding to the first feature data and second identification data corresponding to the second feature data, wherein the first identification data includes first identification parameters corresponding to the plurality of first feature graphs, and the second identification data includes second identification parameters corresponding to the plurality of second feature graphs; The first feature data and the second feature data are identified according to the first identification data and the second identification data, respectively, to generate first converted feature data and second converted feature data.

4. The cooking control method according to claim 3, characterized in that: The acquiring first identification data corresponding to the first feature data and second identification data corresponding to the second feature data includes: acquiring a first marking constant corresponding to the first cooking image data and a second marking constant corresponding to the second cooking image data; Obtaining first identification data according to the first marking constant and the channel positions respectively corresponding to the plurality of first feature maps; Second identification data is obtained according to the second marking constant and the channel positions respectively corresponding to the plurality of second feature maps.

5. The cooking control method according to claim 1, characterized in that: The predicted state data is used to indicate the predicted cooking state category of the target object at the target time, and when the predicted state data satisfies a preset adjustment condition, generating cooking adjustment data based on the predicted state data includes: When the predicted state data indicates that the target object is in a preset cooking state category at a target time and the cooking time corresponding to the target time is less than the preset cooking time, cooking adjustment data is generated based on the predicted state data.

6. The cooking control method according to claim 1, characterized in that: The predicted state data includes predicted cooking state parameters of the target object at a target time, and generating cooking adjustment data based on the predicted state data when the predicted state data satisfies a preset adjustment condition includes: When the degree of deviation of the predicted cooking state parameter from the preset cooking state parameter corresponding to the target time is within a preset range, cooking adjustment data is generated based on the predicted cooking state parameter.

7. The cooking control method according to claim 1, characterized in that: The performing state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at the target time includes: Using the state prediction model to perform state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target time; The method further comprises: According to the sample data, the initial prediction model is trained based on back propagation to obtain the state prediction model. The loss parameter of the back propagation is obtained according to the predicted state loss and the binary cross entropy loss between the predicted state data and the corresponding actual state data. The actual state data is used to indicate the cooking state of the target object at the moment corresponding to the predicted state data.

8. The cooking control method according to claim 1, characterized in that: Also includes: According to the second cooking image data, obtaining the real-time state data corresponding to the target object at the current moment; When the predicted status data and the real-time status data meet the preset alarm conditions, alarm information is generated.

9. A cooking control device, characterized in that: include: An image data acquisition module, configured to acquire first cooking image data and second cooking image data corresponding to a target object, wherein the first cooking image data is obtained by acquiring an image of the target object at a preset historical moment, and the second cooking image data is obtained by acquiring an image of the target object at a current moment, and there is a preset time interval between the preset historical moment and the current moment; a state prediction module, configured to perform state prediction processing based on the first cooking image data and the second cooking image data to obtain predicted state data corresponding to the target object at a target time, wherein the target time is a time after the current time and having a preset time interval with the current time; The cooking adjustment module is used to generate cooking adjustment data based on the predicted state data when the predicted state data meets the preset adjustment conditions, and the cooking adjustment data is used to adjust the cooking control parameters.

10. A cooking device, characterized in that: It includes a cooking component and at least one processor, and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the at least one processor is used to control the cooking component to perform cooking, and the at least one processor implements the cooking control method as described in any one of claims 1 to 8 by executing the instructions stored in the memory.

Images