Combustion pre-warning method and system for a hob

Abstract

The present application relates to the field of cooking equipment, and provides a combustion early warning method and system for a stove, the method comprising: inputting a target image of a target cooking zone into a target recognition model to obtain first coordinates of a stove in the target cooking zone output by the target recognition model; performing cluster analysis on the first coordinates to obtain second coordinates of the stove; in a case where the stove is in a fire-on state, obtaining an area intersection ratio of a pot and the stove based on real-time image information of the target cooking zone and the second coordinates of the stove; in a case where the area intersection ratio is less than a target threshold, outputting an empty burning early warning instruction, and / or outputting a control instruction for reducing the firepower of the stove; wherein the target image is collected by a camera of a cooking equipment, and the target recognition model is trained by taking sample images as samples and taking pre-determined coordinates of the stove corresponding to the sample images as sample labels. The method improves the safety and intelligence of the stove.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cooking equipment, in particular to a combustion early warning method and system of a cooking appliance. BACKGROUND

[0002] With the improvement of living standards, people have higher and higher requirements for food, and more and more people will use their free time to cook to satisfy their love of food.

[0003] In the prior art, the combustion is early warned by detecting and identifying the flame image on the cooking appliance, but in the actual cooking process, due to the fact that the flame on the cooking appliance has a lower recognition degree than the burning flame in a natural scene, and the influence of the cooking environment light, the accuracy of the flame detection and identification of the cooking appliance is reduced, the combustion cannot be accurately early warned, and the safety of the cooking equipment is greatly reduced. SUMMARY

[0004] The present application aims to at least solve one of the technical problems existing in the prior art. To this end, the present application provides a combustion early warning method of a cooking appliance to improve the safety of the cooking appliance.

[0005] The combustion early warning method of the cooking appliance according to the first aspect of the present application comprises:

[0006] inputting a target image of a target cooking zone into a target recognition model to obtain a first coordinate of a cooking appliance in the target cooking zone output by the target recognition model;

[0007] performing cluster analysis on the first coordinate to obtain a second coordinate of the cooking appliance;

[0008] in a case where the cooking appliance is in a fire-on state, obtaining an area intersection ratio of a pot and the cooking appliance based on real-time image information of the target cooking zone and the second coordinate of the cooking appliance;

[0009] in a case where the area intersection ratio is less than a target threshold, outputting an empty burning early warning instruction, and / or outputting a control instruction for reducing the firepower of the cooking appliance;

[0010] wherein the target image is collected by a camera of a cooking equipment, and the target recognition model is trained by taking a sample image as a sample and taking a pre-determined coordinate of a cooking appliance corresponding to the sample image as a sample label.

[0011] The combustion early warning method of the cooking appliance according to the present application calibrates the position information of the cooking appliance on the cooking zone, effectively tracks and detects the cooking process, calculates the area intersection ratio between the pot and the cooking appliance, discovers that the pot moves out of the range of the cooking appliance, provides an alarm prompt or reduces the firepower for the user, avoids the loss caused by the empty burning of the cooking appliance in time, and greatly improves the safety and intelligence of the cooking appliance.

[0012] According to one embodiment of the present application, after the area intersection ratio of the obtained pot and the stove is obtained, the method further comprises:

[0013] In the case that the area intersection ratio is not less than a target threshold value and there is no food in the pot, outputting a dry burning warning instruction, and / or outputting a control instruction for reducing the firepower of the stove.

[0014] According to one embodiment of the present application, the clustering analysis on the first coordinates to obtain the second coordinates of the stove comprises:

[0015] Obtaining the number of stoves, and taking the number of stoves as the cluster number of clustering analysis;

[0016] According to the cluster number, the first coordinates are subjected to clustering analysis to obtain the second coordinates.

[0017] According to one embodiment of the present application, the obtaining of the number of stoves comprises:

[0018] Obtaining the equipment information of the target stove;

[0019] Based on the equipment information, obtaining the number of stoves.

[0020] According to one embodiment of the present application, the obtaining of the number of stoves comprises:

[0021] Inputting the target image of the target stove into the target recognition model to obtain the number of stoves in the target stove output by the target recognition model.

[0022] According to one embodiment of the present application, the clustering analysis on the first coordinates to obtain the second coordinates according to the cluster number comprises:

[0023] According to the cluster number, the first coordinates are divided into first coordinates of different stoves;

[0024] Obtaining the first coordinates of two opposite corners in the first coordinates of different stoves to obtain the center point coordinates of the stove;

[0025] Performing clustering analysis on the center point coordinates to obtain target center point coordinates;

[0026] Calculating the distance values between each point in the first coordinates of the stove and the target center point coordinates;

[0027] Comparing the distance values with a distance threshold value, and discarding the points corresponding to the distance values greater than the distance threshold value;

[0028] Reserve the distance value not greater than the distance threshold value corresponding to the point, obtain the second coordinate.

[0029] The cooking pre-warning system of the stove according to the second aspect of the present application comprises:

[0030] A cooking device;

[0031] A camera mounted on the cooking device;

[0032] A target cooking zone provided with a stove, the target cooking zone being located in a shooting area of the camera;

[0033] A controller electrically connected with the camera, the controller being configured to output an empty-burning pre-warning instruction and / or a control instruction for reducing the firepower of the stove based on any one of the cooking pre-warning methods of the stove described above when the area intersection ratio is less than a target threshold.

[0034] According to one embodiment of the present application, the controller is further configured to output a dry-burning pre-warning instruction and / or reduce the firepower of the stove when the area intersection ratio is not less than a target threshold and there is no food in the pot.

[0035] The cooking pre-warning device of the stove according to the third aspect of the present application comprises:

[0036] A first processing module configured to input a target image of a target cooking zone into a target recognition model to obtain first coordinates of a stove in the target cooking zone output by the target recognition model;

[0037] A second processing module configured to perform cluster analysis on the first coordinates to obtain second coordinates of the stove;

[0038] A calculation module configured to, when the stove is in a fire-on state, obtain an area intersection ratio of a pot and the stove based on real-time image information of the target cooking zone and the second coordinates of the stove;

[0039] A control module configured to output an empty-burning pre-warning instruction and / or a control instruction for reducing the firepower of the stove when the area intersection ratio is less than a target threshold;

[0040] The target image is collected by a camera of a cooking device, and the target recognition model is trained by taking a sample image as a sample and taking pre-determined coordinates of a stove corresponding to the sample image as a sample label.

[0041] The electronic device according to the fourth aspect of the present application comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the combustion early warning method of the stove according to any one of the above aspects when executing the computer program.

[0042] The non-transitory computer readable storage medium according to the fifth aspect of the present application stores a computer program, and the computer program implements the steps of the combustion early warning method of the stove according to any one of the above aspects when executed by a processor.

[0043] The computer program product according to the sixth aspect of the present application comprises a computer program, and the computer program implements the steps of the combustion early warning method of the stove according to any one of the above aspects when executed by a processor.

[0044] The one or more technical solutions described above in the embodiments of the present application have at least one of the following technical effects:

[0045] By calibrating the position information of the stove on the cooking bench, the cooking process is effectively tracked and detected, the combustion early warning of the stove is realized, and the safety and intelligence of the stove are improved.

[0046] Further, the accuracy of the stove position is improved by deep learning combined with clustering analysis, the accuracy of the area intersection ratio calculation is improved, and the accuracy of the dry burning early warning and the empty burning early warning is ensured.

[0047] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0049] Figure 1 is a flowchart of the combustion early warning method of the stove provided by the embodiments of the present application;

[0050] Figure 2 is a structural schematic diagram of the stove provided by the embodiments of the present application;

[0051] Figure 3 is one of the target cooking bench schematic diagrams provided by the embodiments of the present application;

[0052] Figure 4 is another target cooking bench schematic diagram provided by the embodiments of the present application;

[0053] Figure 5 is a structural schematic view of a combustion pre-warning device of a stove provided by an embodiment of the present application;

[0054] Figure 6 is a structural schematic view of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0055] The embodiments of the present application will be further described below in conjunction with the accompanying drawings and embodiments. The following embodiments are used to illustrate the present application, but cannot be used to limit the scope of the present application.

[0056] In the description of the embodiments of the present application, it should be noted that the terms “center”, “longitudinal”, “transverse”, “upper”, “lower”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inner”, “outer” and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the embodiments of the present application and simplify the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the embodiments of the present application. In addition, the terms “first”, “second”, “third” are only for descriptive purposes, and cannot be understood as indicating or implying relative importance.

[0057] In the description of the embodiments of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms “connected” and “connected” should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or it can be integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium. For those skilled in the art, the specific meaning of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0058] In the embodiments of the present application, unless otherwise explicitly specified and limited, the first feature is “on” or “under” the second feature, which can be that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, the first feature can be directly above or obliquely above the second feature, or it can only mean that the horizontal height of the first feature is higher than that of the second feature. The first feature can be directly below or obliquely below the second feature, or it can only mean that the horizontal height of the first feature is less than that of the second feature.

[0059] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the embodiments of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any suitable manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0060] The following will be described in combination with Figures 1 to 4 The method for predicting the combustion of the stove of the embodiments of the present application is described, and the execution subject of the method can be the controller of the device end, or the cloud, or the edge server.

[0061] As Figure 1 The method for predicting the combustion of the stove of the present application includes steps 110 to 140.

[0062] Step 110, input the target image of the target stove into the target recognition model, and obtain the first coordinate of the stove in the target stove.

[0063] By installing a camera on the cooking equipment to shoot the target image of the target stove, the pot, stove, food and cooking behavior in the target image can be detected and recognized, and then the recognized information can be used to assist the user in cooking.

[0064] The cooking equipment can be a hood used with the target stove, and the cooking equipment can also be the target stove. The camera is installed at a position where it can shoot all the stoves of the target stove.

[0065] The target recognition model is constructed by a target detection algorithm based on a deep learning framework, and the target recognition model can recognize and output the coordinates of the stove and the stove in the target image.

[0066] As Figure 2 The camera 10 is installed on the hood 30, and the camera 10 can shoot the target image of the target stove 20. The target image can be input into the target recognition model, and the first coordinates of the first stove 21 and the second stove 22 on the target stove 20 can be input. The first switch 23 controls the switch and fire adjustment of the first stove 21, and the second switch 24 controls the switch and fire adjustment of the second stove 22.

[0067] In actual execution, the target recognition model needs to be trained first, and then the target image of the target stove shot by the camera is input into the target recognition model for recognition, and the coordinates of the stove are output.

[0068] The training of the target recognition model takes a sample image as a sample, and a predetermined coordinate of the cooking bench corresponding to the sample image as a sample label.

[0069] The target detection algorithm for constructing the target recognition model can be selected according to the requirements of the use scene and the computing hardware. The target detection algorithm can be an anchor-based algorithm, such as Yolo, SSD, RetinaNet, FasterRCNN, and other single-stage or double-stage detection algorithms. The target detection algorithm can also be an anchor-free algorithm, such as CornerNet, CenterNet, and FCOS.

[0070] When the target recognition model is deployed in an embedded chip, a lightweight recognition and detection model such as mobilenet, shufflenet, and ghostnet can be selected.

[0071] The first coordinate output by the target recognition model, wherein the first coordinate is the initial coordinate of the cooking appliance and includes some coordinates that do not belong to the cooking appliance.

[0072] It can be understood that by taking multiple images of the target cooking bench with the camera, multiple image information of the target cooking bench is obtained. The more the number of image information, the higher the accuracy of the first coordinate of the cooking appliance output by the target recognition model.

[0073] In actual execution, each image of the target cooking bench can be input into the trained target recognition model to output the first coordinate of the cooking appliance respectively. Multiple images of the target cooking bench can also be input into the trained target recognition model in batches to output the first coordinate of the cooking appliance respectively.

[0074] It can be understood that multiple cooking appliances can be provided on the target cooking bench, and the target recognition model identifies the target image to output the first coordinate of all the cooking appliances on the target cooking bench.

[0075] Step 120, clustering analysis is performed on the first coordinate to obtain the second coordinate of the cooking appliance.

[0076] In this step, the first coordinate of the cooking appliance is subjected to clustering analysis, and the clustering algorithm is used to process the first coordinate to discard the coordinates of the points that do not belong to the corresponding cooking appliance in the first coordinate, thereby obtaining the second coordinate of the cooking appliance.

[0077] The first coordinate and the second coordinate are a set of coordinates of each point in the cooking appliance of the target image, and the second coordinate is obtained by removing the coordinate points that do not belong to the corresponding cooking appliance after clustering processing of the first coordinate.

[0078] In actual execution, the first coordinates can be subjected to cluster analysis by a clustering algorithm such as a K-MEANS algorithm, a K-MEDOIDS algorithm, a CLARANS algorithm, and the like.

[0079] It can be understood that the more the number of target cooking table images subjected to cluster analysis of the first coordinates, the more accurate the second coordinates obtained.

[0080] Step 130, based on the real-time image information of the target cooking table and the second coordinates of the cooking appliance, the area intersection ratio of the pot and the cooking appliance is obtained when the cooking appliance is in the fire-on state.

[0081] The user controls the cooking appliance switch and the fire adjustment through the cooking appliance switch, and can set a sensor at the cooking appliance switch or directly set a sensor at the cooking appliance to determine whether the cooking appliance is in the fire-on state.

[0082] It can be understood that the target recognition model can be trained accordingly, and whether the cooking appliance is in the fire-on state can be determined from the real-time image of the target cooking table captured by the camera.

[0083] The training of the target recognition model can use sample images as samples and use the pre-determined combustion state information of the sample image corresponding to the sample image as a sample label for training.

[0084] When the cooking appliance is in the fire-on state, the real-time image information of the target cooking table is obtained in real time through the camera, and the information of the pot, the cooking appliance, the food material and the cooking behavior on the target cooking table is obtained according to the real-time image information of the target cooking table.

[0085] In this step, based on the real-time image information of the target cooking table, the specific position of the pot on the target cooking table is obtained, based on the second coordinates of the cooking appliance, the position of the cooking appliance on the target cooking table is determined, and then whether the pot is on the cooking appliance is determined.

[0086] The position of the cooking appliance on the target cooking table is fixed, and when the cooking appliance is in the fire-on state, the position of the pot will change when the user performs cooking behaviors such as stir-frying, adding ingredients and pouring oil.

[0087] When the position of the pot overlaps with the position of the cooking appliance in the fire-on state, the pot is on the cooking appliance for heating, and when the position of the pot is completely separated from the position of the cooking appliance, the pot is away from the cooking appliance, and the cooking appliance is in the empty burning state.

[0088] The area intersection ratio of the target cooking table is the proportion of the area of the overlapping part of the pot and the cooking appliance to the area of the cooking appliance, and the area intersection ratio can be calculated by the position information of the pot and the second coordinates of the cooking appliance.

[0089] Step 140: If the area intersection ratio is less than the target threshold, output an empty burning warning command and / or a control command to reduce the stove's firepower.

[0090] When the stove is on, real-time image information of the target stove is acquired through the camera. Based on the real-time image information of the target stove, the position information of the pot on the target stove is obtained. Combined with the second coordinate of the stove, the area intersection ratio is calculated.

[0091] The area intersection-union ratio is compared with the target threshold. When the area intersection-union ratio is less than the target threshold, the pot is far away from the stove when the pot is completely separated from the stove. When the area intersection-union ratio is not less than the target threshold, the pot is on the stove when the pot is on top of the stove.

[0092] When the area overlap ratio is less than the target threshold while the stove is on, the cookware is not on the stove and the stove is in a state of dry burning, which may cause a fire or burn the user.

[0093] When the area overlap ratio is less than the target threshold while the stove is on, it can output a dry-burning warning command, control the cooking equipment to alarm or the individual alarm to alarm, so as to warn the user; it can also control the stove to reduce the firepower to reduce the risk of fire or burns to the user; it can also control the stove to reduce the firepower at the same time as outputting the dry-burning warning command.

[0094] The relevant technology uses image detection to detect flame information and trigger an alarm. However, in actual cooking, the flames observed in the images are not as easily recognizable as flames in natural scenes. Furthermore, the accuracy of the flame detection is further reduced due to the influence of the lighting environment.

[0095] The combustion warning method for stoves of the present invention effectively tracks and detects the cooking process by calibrating the position information of the stove on the stovetop, calculating the area intersection ratio between the pot and the stove, and providing an alarm prompt or reducing the firepower when the pot moves out of the range of the stove, so as to avoid losses caused by the stove burning dry, and greatly improve the safety and intelligence of the stove.

[0096] The combustion warning method for stoves provided by this invention determines the position information of the stove on the target stove through deep learning and cluster analysis, and performs combustion warning based on the area ratio of the stove and the pot when the stove is on, which greatly improves the safety and intelligence of the stove.

[0097] In some embodiments, after step 130, the area intersection ratio of the stove and the pot is obtained in real time and compared with a target threshold. When the area intersection ratio is not less than the target threshold and no food is detected in the pot, a dry burning warning is issued.

[0098] When the cooking area is not less than the target threshold, and the cookware is heated on the stove, if no food is detected in the cookware, a dry-burning warning command can be output to control the cooking equipment to alarm or a separate alarm to alert the user; it can also output a control command to control the stove to reduce the firepower, reduce dry-burning damage to the cookware, and avoid burning the user; or it can output a control command at the same time as outputting the dry-burning warning command to control the stove to reduce the firepower.

[0099] In actual operation, it was found that when the user turned on the stove and the stove was in the ignition state, the camera would capture real-time image information of the target stove and obtain the information of the food in the pot on the target stove based on the real-time image information of the target stove.

[0100] Understandably, the target recognition model can be trained accordingly to determine whether there is food in the pot, the type of food, and the weight of the food from the real-time image of the target stove captured by the camera.

[0101] It should be noted that the food information inside the cookware includes information on the oil, water, vegetables, and meat inside the cookware.

[0102] In some embodiments, step 120 performs cluster analysis, which can use the number of stoves as the number of clusters in the cluster analysis, and process the first coordinates of the stoves to obtain the second coordinates of the stoves.

[0103] The number of stoves on the target stove is used as the number of clusters for cluster analysis. The first coordinates of multiple stoves on the same target stove are divided into different clusters, processed, and the corresponding second coordinates are obtained by clustering.

[0104] Using the number of stoves on the target stove as the cluster number, cluster analysis can effectively distinguish the first coordinates of different stoves, improve the accuracy of the obtained second coordinates, and thus improve the accuracy of the calculated area intersection ratio, thereby enhancing the safety of the stoves.

[0105] like Figure 3 As shown, there are two stoves on the target stove. Taking each stove as a stove, when analyzing the first coordinate of the stove, the number of stoves is used as the cluster number to divide the first coordinate into the first coordinates of different stoves.

[0106] like Figure 4 As shown, after performing cluster analysis on the first coordinates of different stoves, the second coordinates of different stoves are obtained. For example, the second coordinates of the first stove 21 and the second stove 22 can also be obtained, as well as the second coordinates of stove switches such as the first switch 23 and the second switch 24.

[0107] In some embodiments, the number of stoves on the target stove can be obtained through the device information of the target stove or the output of the target recognition model.

[0108] First, based on the equipment information of the target stove, the number of stoves is obtained.

[0109] In practice, device information can be obtained by reading the model number of the target stove, or by having the user input the model number of the target stove and downloading the corresponding device information from the cloud, or by having the user directly input the device information.

[0110] The equipment information may include the type and number of stoves, switches, or brand logos on the target stove.

[0111] Secondly, the target image of the target stove is input into the target recognition model, and the target recognition model outputs the number of stoves.

[0112] A target recognition model is constructed using a target detection algorithm based on a deep learning framework. This model can identify stoves in target images and output the type and number of stoves.

[0113] In practice, the target recognition model can be trained using sample images as samples and pre-determined types of sample objects corresponding to the sample images as sample labels.

[0114] Understandably, by taking multiple images of the target stove with a camera, multiple image information of the target stove can be obtained. The more image information there is, the higher the accuracy of the target recognition model in outputting the type and number of stoves.

[0115] In practice, the number of stoves output by each target stove image can be counted after inputting into the trained target recognition model, and the number of stoves that appear most frequently can be selected as the number of clusters for cluster analysis.

[0116] In some embodiments, after determining the number of clusters based on the number of stoves, a cluster analysis is performed on the first coordinates of the stoves based on the number of clusters to obtain the second coordinates.

[0117] In this embodiment, the first coordinates of multiple stoves are divided according to the number of clusters determined by the number of stoves, and the first coordinates of the stoves are divided into different first coordinates.

[0118] For a single stove, obtain the first coordinates of the two diagonal points in the first coordinate system of the stove, calculate the average of the first coordinates of the two diagonal points, and obtain the coordinates of the center point of the stove.

[0119] For multiple stoves on the target stove, the coordinates of the center point of each stove are calculated based on the first coordinates of the two diagonal points of each stove.

[0120] In actual implementation, multiple target stove images are captured by cameras, which can obtain the coordinates of multiple center points of the same stove. Cluster analysis is then performed on the coordinates of the multiple center points of the stove to obtain the coordinates of the target center point.

[0121] The distance between the coordinates of each point in the first coordinate system and the coordinates of the target center point is calculated. Using this distance as a criterion, the points in the first coordinate system are clustered, and valid points are selected and retained while invalid points are discarded.

[0122] Based on the principle that objects within the same cluster have high similarity, while objects in different clusters have low similarity, the distance value is compared with a preset distance threshold. If the distance value calculated for a point in the first coordinate is greater than the distance threshold, it indicates that the point has low similarity with objects in its cluster, and the coordinate corresponding to that point is discarded.

[0123] If the distance value calculated for a point in the first coordinate system is not greater than the distance threshold, it indicates that the point has a high similarity to objects within the same cluster, and the coordinates corresponding to that point are retained.

[0124] The distance between the coordinates of all points in the first coordinate system and the coordinates of the target center point is calculated and compared with a preset distance threshold. The retained first coordinate system is the second coordinate system of the stove.

[0125] The following is a specific example.

[0126] The target stove is a set of two stoves. Sixteen target images of the target stove are obtained through a camera. These sixteen target images are then input into a target recognition model to obtain the first coordinates of the two stoves.

[0127] For a single stove, the first coordinates of 16 points with two opposite corners were obtained statistically. in, and The top-left x-coordinate, top-left y-coordinate, bottom-right x-coordinate, and bottom-right y-coordinate represent the position coordinates of the i-th stove, where i = 1...n. This represents the n stoves detected in all 16 images, where n is 2.

[0128] The top left and bottom right points are the diagonal points of a single stove.

[0129] For each stove location Calculate the coordinates of the center point of each stove.

[0130] For 16 target images, the coordinates of 16 center points for each stove are calculated, resulting in a total of 32 center point coordinates for the two stoves. Cluster analysis is then performed on these 32 center point coordinates to obtain the target center point coordinates for each stove after clustering, resulting in a total of two target center point coordinates.

[0131] Based on the coordinates of the target center point, the first coordinates of the stove in each target image are grouped. A preset distance threshold is set. If the distance between the first coordinate of a point of the stove in the group and the coordinates of the target center point is greater than the distance threshold, the first coordinate of that point is discarded. If the distance is not greater than the distance threshold, the first coordinate of that point is retained.

[0132] The retained first coordinates are used as the second coordinates of the stove. The average value of the center coordinates of the stove is recalculated based on the second coordinates, and then the size parameters of the stove are calculated, such as the average width and height of the stove frame. This accurately determines the position of the stove on the target stovetop and improves the accuracy of subsequent visual recognition tasks, which may include tasks such as recognizing pots, ingredients and cooking behavior.

[0133] The present invention also provides a combustion warning system for a stove, comprising: cooking equipment, camera, target stove and controller.

[0134] The camera is installed on the cooking equipment to capture a target image of the target stove.

[0135] The cooking equipment can be a range hood used in conjunction with the target stove, or it can be the target stove itself. The target stove is located within the camera's field of view, and the camera is installed in a position that can capture all the cooktops on the target stove.

[0136] After the controller is electrically connected to the camera, it can output corresponding combustion warning commands or control commands to control the firepower of the stove based on the area overlap ratio of the overlapping parts of the pot and the stove, using the aforementioned stove combustion warning method.

[0137] Based on the real-time image information of the target stove, the specific position of the pot on the target stove is obtained. Based on the second coordinate of the stove, the position of the stove on the target stove is determined, and then it is determined whether the pot is on the stove.

[0138] When the stove is on, the controller acquires real-time image information of the target stove through the camera. Based on the real-time image information of the target stove, it obtains the position information of the pot on the target stove and calculates the area intersection ratio by combining the second coordinate of the stove.

[0139] The controller compares the area intersection ratio with the target threshold. When the area intersection ratio is less than the target threshold, the pot is completely separated from the stove and is away from the stove. When the area intersection ratio is not less than the target threshold, the pot is overlapped with the stove and is on the stove.

[0140] When the area overlap ratio is less than the target threshold while the stove is on, the cookware is not on the stove and the stove is in a state of dry burning, which may cause a fire or burn the user.

[0141] When the controller detects that the area crossover ratio is less than the target threshold while the flame is on, it can output a dry-burning warning command to control the cooking equipment to alarm or a separate alarm to alert the user; it can also output a control command to control the stove to reduce the firepower and reduce the risk of fire or burns to the user.

[0142] The combustion warning system for stoves provided by this invention determines the location information of the stove on the target stove through deep learning and cluster analysis. Based on the area ratio of the stove and the pot when the stove is on, a combustion warning is given, which greatly improves the safety and intelligence of the stove.

[0143] In some embodiments, the controller calculates the area intersection ratio of the stove and the pot in real time and compares it with a target threshold. When the area intersection ratio is not less than the target threshold and no food is detected in the pot, a dry burning warning is issued.

[0144] When the fire is on, if the controller detects that the area cross-section ratio is not less than the target threshold, the cookware is being heated on the stove, and there is no food in the cookware, it can output a dry-burning warning command to control the cooking equipment to alarm or a separate alarm to alert the user; it can also output a control command to control the stove to reduce the firepower, reduce dry-burning damage to the cookware, and avoid burning the user.

[0145] The combustion warning device for a stove provided in the embodiments of the present invention will be described below. The combustion warning device for a stove described below can be referred to in correspondence with the combustion warning method for a stove described above.

[0146] like Figure 5 As shown, the combustion warning device for a stove provided in this embodiment of the invention includes:

[0147] The first processing module 510 is used to input the target image of the target stove into the target recognition model and obtain the first coordinates of the stove in the target stove output by the target recognition model.

[0148] The second processing module 520 is used to perform cluster analysis on the first coordinates to obtain the second coordinates of the stove;

[0149] The calculation module 530 is used to obtain the area intersection ratio of the pot and the stove based on the real-time image information of the target stove and the second coordinate of the stove when the stove is in the ignition state.

[0150] The control module 540 is used to output an empty-burning warning command when the area intersection ratio is less than the target threshold, and / or output a control command to reduce the firepower of the stove.

[0151] The target image is captured by the camera of the cooking equipment. The target recognition model is trained using the sample image as the sample and the pre-determined coordinates of the stove corresponding to the sample image as the sample label.

[0152] The combustion warning device for stoves provided by the present invention determines the position information of the stove on the target stove through deep learning and cluster analysis, and performs combustion warning based on the area ratio of the stove and the pot when the stove is on, which greatly improves the safety and intelligence of the stove.

[0153] In some embodiments, after the calculation module 530 calculates the area intersection ratio of the cookware and the stove, the control module 540 is further configured to output a dry burning warning command and / or output a control command to reduce the firepower of the stove when the area intersection ratio is not less than the target threshold and there is no food in the cookware.

[0154] In some embodiments, the second processing module 520 is used to obtain the number of stoves, use the number of stoves as the number of clusters for cluster analysis, and perform cluster analysis on the first coordinates based on the number of clusters to obtain the second coordinates.

[0155] In some embodiments, the second processing module 520 obtains the number of stoves by: obtaining device information of the target stove; and obtaining the number of stoves based on the device information.

[0156] In some embodiments, the second processing module 520 obtains the number of stoves by: inputting the target image of the target stove into the target recognition model, and obtaining the number of stoves in the target stove output by the target recognition model.

[0157] In some embodiments, the second processing module 520 is used to divide the first coordinates into the first coordinates of different stoves according to the number of clusters; obtain the first coordinates of two diagonal points in the first coordinates of different stoves, and calculate the center point coordinates of the stoves; perform cluster analysis on the center point coordinates to obtain the target center point coordinates; calculate the distance value between each point in the first coordinates of the stoves and the target center point coordinates; compare the distance value with a distance threshold, discard the points whose distance value is greater than the distance threshold; and retain the points whose distance value is not greater than the distance threshold to obtain the second coordinates.

[0158] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 can call the logical instructions in the memory 630 to execute a combustion warning method for the stove. The method includes: inputting a target image of the target stove into a target recognition model to obtain the first coordinates of the stove in the target stove output by the target recognition model; performing cluster analysis on the first coordinates to obtain the second coordinates of the stove; when the stove is in the ignition state, obtaining the area intersection-union ratio (IU / R) of the pot and the stove based on the real-time image information of the target stove and the second coordinates of the stove; when the IU / R is less than a target threshold, outputting a dry-burning warning command, and / or outputting a control command to reduce the stove's firepower; wherein the target image is acquired by a camera of the cooking device, and the target recognition model is trained using sample images as samples and pre-determined coordinates of the stove corresponding to the sample images as sample labels.

[0159] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0160] Furthermore, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the stove combustion warning method provided in the above-described method embodiments. The method includes: inputting a target image of the target stove into a target recognition model to obtain the first coordinates of the stove in the target stove output by the target recognition model; performing cluster analysis on the first coordinates to obtain the second coordinates of the stove; when the stove is in the ignition state, obtaining the area intersection-union ratio (IU / R) of the pot and the stove based on the real-time image information of the target stove and the second coordinates of the stove; when the IU / R is less than a target threshold, outputting a dry-burning warning command, and / or outputting a control command to reduce the stove's firepower; wherein the target image is acquired by a camera of the cooking device, and the target recognition model is trained using sample images as samples and pre-determined stove coordinates corresponding to the sample images as sample labels.

[0161] On the other hand, embodiments of the present invention also provide a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the combustion warning method for a stove provided in the above embodiments. The method includes: inputting a target image of a target stove into a target recognition model to obtain the first coordinates of the stove in the target stove output by the target recognition model; performing cluster analysis on the first coordinates to obtain the second coordinates of the stove; when the stove is in the ignition state, obtaining the area intersection-union ratio (IU / R) of the pot and the stove based on the real-time image information of the target stove and the second coordinates of the stove; when the IU / R is less than a target threshold, outputting a dry-burning warning command, and / or outputting a control command to reduce the firepower of the stove; wherein the target image is acquired by a camera of a cooking device, and the target recognition model is trained using sample images as samples and pre-determined coordinates of the stove corresponding to the sample images as sample labels.

[0162] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0163] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A combustion early warning method for a stove, characterized in that, include: The target image of the target stove is input into the target recognition model to obtain the first coordinates of the stove in the target stove output by the target recognition model; Cluster analysis is performed on the first coordinates to obtain the second coordinates of the stove; the second coordinates are obtained by removing coordinate points that do not belong to the corresponding stove after clustering the first coordinates. When the stove is in the ignition state, the area intersection ratio of the pot and the stove is obtained based on the real-time image information of the target stove and the second coordinates of the stove; If the area intersection ratio is less than the target threshold, an empty burning warning command is output, and / or a control command for reducing the firepower of the stove is output; The target image is captured by the camera of the cooking equipment and is taken of the target stove. The target recognition model is trained by using the sample image as a sample and the pre-determined coordinates of the stove corresponding to the sample image as the sample label. After obtaining the area intersection ratio of the cookware and the stove, the method further includes: If the area intersection ratio is not less than the target threshold and there is no food in the pot, output a dry burning warning command and / or output a control command to reduce the firepower of the stove.

2. The combustion early warning method for a stove according to claim 1, characterized in that, The step of performing cluster analysis on the first coordinates to obtain the second coordinates of the stove includes: The number of stoves is obtained, and the number of stoves is used as the number of clusters for cluster analysis; Based on the number of clusters, the first coordinates are subjected to cluster analysis to obtain the second coordinates.

3. The combustion early warning method for a stove according to claim 2, characterized in that, The process of obtaining the number of stoves includes: Obtain the device information of the target stove; Based on the device information, the number of stoves is obtained.

4. The combustion early warning method for a stove according to claim 2, characterized in that, The process of obtaining the number of stoves includes: The target image of the target stove is input into the target recognition model to obtain the number of stoves in the target stove as output by the target recognition model.

5. The combustion early warning method for a stove according to claim 2, characterized in that, The step of performing cluster analysis on the first coordinates based on the number of clusters to obtain the second coordinates includes: Based on the number of clusters, the first coordinates are divided into the first coordinates of different stoves; By obtaining the first coordinates of two diagonal points in the first coordinates of different stoves, the coordinates of the center point of the stove are obtained; Cluster analysis is performed on the coordinates of the center point to obtain the coordinates of the target center point; Calculate the distance between each point in the first coordinate system of the stove and the coordinate system of the target center point; The distance value is compared with a distance threshold, and points whose distance value is greater than the distance threshold are discarded. The points whose distance values ​​are not greater than the distance threshold are retained to obtain the second coordinates.

6. A combustion early warning system for a stove, characterized in that, include: Cooking equipment; A camera is mounted on the cooking device; The target stove is equipped with a stove and is located within the shooting area of ​​the camera. A controller electrically connected to the camera, the controller being configured to, based on the combustion warning method for a stove according to any one of claims 1-5, output an empty-burning warning command when the area intersection ratio is less than a target threshold, and / or output a control command for reducing the firepower of the stove; The controller is also configured to output a dry-burning warning command and / or output a control command to reduce the firepower of the stove when the area intersection ratio is not less than the target threshold and there is no food in the pot.

7. A combustion early warning device for a stove, characterized in that, include: The first processing module is used to input the target image of the target stove into the target recognition model and obtain the first coordinates of the stove in the target stove output by the target recognition model; The second processing module is used to perform cluster analysis on the first coordinates to obtain the second coordinates of the stove; the second coordinates are obtained by removing coordinate points that do not belong to the corresponding stove after performing cluster processing on the first coordinates. The calculation module is used to obtain the area intersection ratio of the pot and the stove based on the real-time image information of the target stove and the second coordinates of the stove when the stove is in the ignition state. The control module is configured to output an empty-burning warning command when the area intersection ratio is less than the target threshold, and / or output a control command to reduce the firepower of the stove. The target image is captured by the camera of the cooking equipment and is taken of the target stove. The target recognition model is trained by using the sample image as a sample and the pre-determined coordinates of the stove corresponding to the sample image as the sample label. After the calculation module calculates the area intersection ratio of the pot and the stove, the control module is further used for: If the area intersection ratio is not less than the target threshold and there is no food in the pot, output a dry burning warning command and / or output a control command to reduce the firepower of the stove.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the combustion warning method for the stove as described in any one of claims 1 to 5.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the combustion warning method for the stove as described in any one of claims 1 to 5.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the combustion warning method for the stove as described in any one of claims 1 to 5.

Images