Overflow detection method and system, range hood and storage medium

Abstract

The present invention provides a pot overflow detection method and system, a range hood and a storage medium. The pot overflow detection method comprises: obtaining point cloud data of a first area within a first preset distance range from the edge of a pot in real time; obtaining point cloud change data of each point cloud in a preset period based on the point cloud data of the first area; if the point cloud change data reaches the pot overflow threshold, generating a pot overflow control signal to turn off the stove fire source. The present invention obtains point cloud change data of each point cloud in a preset period through the point cloud data of the first area within a first preset distance range from the edge of the pot. If the point cloud change data reaches the pot overflow threshold, generating a pot overflow control signal to turn off the stove fire source, that is, judging whether there is a pot overflow by the change of the point cloud at the edge of the pot, and turning off the stove fire source when the pot overflows.

Description

Technical Field

[0001] The present invention relates to the technical field of kitchenware, and in particular to a pan overflow detection method and system, a range hood and a storage medium. Background Art

[0002] There are currently three methods for detecting overflow:

[0003] 1) Design a pot with overflow detection function or design an adaptive pot cover to avoid overflow. Disadvantages: difficult to use in real life, strange shape and cannot interact with users.

[0004] 2) The stove is equipped with a temperature probe that can detect overflowing pots through temperature changes. Disadvantage: Smart stoves that detect overflowing pots through temperature changes are prone to misjudgment based solely on temperature information.

[0005] 3) Install a camera on the range hood to collect images of the stove to monitor whether the pot is overflowing. Disadvantage: Using a camera to collect images will involve user privacy.

[0006] The above three methods for detecting overflowing pot each have disadvantages. Summary of the invention

[0007] The technical problem to be solved by the present invention is to overcome the defects of the methods for detecting overflowing pot in the prior art, and to provide an overflowing pot detection method and system, a range hood and a storage medium.

[0008] The present invention solves the above technical problems through the following technical solutions:

[0009] The present invention provides a pot overflow detection method, the pot overflow detection method comprising:

[0010] Acquire point cloud data of a first area within a first preset distance range from the edge of the pot in real time;

[0011] Obtaining point cloud change data of each point cloud in a preset period based on the point cloud data of the first area;

[0012] If the point cloud change data reaches the pot overflow threshold, an overflow control signal is generated to turn off the fire source of the stove.

[0013] Preferably, before the step of acquiring point cloud data of the first area within a preset distance range from the pot edge in real time, the pot overflow detection method further comprises:

[0014] Acquire total area point cloud data of a total detection area, where the total detection area covers a stove surface;

[0015] Obtaining a grayscale image containing the pot based on the total area point cloud data;

[0016] Identify the grayscale image to obtain the position of the pot cover;

[0017] The step of acquiring point cloud data of a first area within a first preset distance from the edge of the pot in real time specifically comprises:

[0018] The first area point cloud data is extracted from the total area point cloud data according to the position of the pot cover.

[0019] Preferably, the step of extracting the first area point cloud data from the total area point cloud data according to the position of the pot cover specifically includes:

[0020] Extracting second area point cloud data within a second preset distance from the edge of the pot cover from the total area point cloud data according to the position of the pot cover;

[0021] Obtaining a plan view including the pot based on the second area point cloud data;

[0022] Determine the edge position of the pot by extracting arcs based on the plan view;

[0023] The first area point cloud data is extracted from the total area point cloud data according to the edge position.

[0024] Preferably, the step of extracting the first area point cloud data from the total area point cloud data according to the edge position specifically includes:

[0025] The point cloud data of the first area is obtained by extracting point cloud data within a first preset distance range from the edge position from the point cloud data of the total area based on a plane fast search algorithm.

[0026] Preferably, the step of obtaining the total area point cloud data of the total detection area specifically includes:

[0027] Obtain point cloud data of the total detection area in the reference coordinate system of the shooting device;

[0028] The total area point cloud data in the geodetic coordinate system is obtained according to the conversion relationship between the shooting device reference coordinate system and the geodetic coordinate system.

[0029] Preferably, the first area point cloud data includes the number of point clouds, and the overflow detection method further includes:

[0030] Get the number of initial point clouds corresponding to the initial time point;

[0031] If the point cloud change data reaches the pot overflow threshold, the step of generating a pot overflow control signal to turn off the stove fire source specifically includes:

[0032] The point cloud quantity difference between the current point cloud quantity and the initial point cloud quantity is calculated, and if the point cloud quantity difference reaches the pot overflow threshold, an overflow control signal is generated to turn off the fire source of the stove.

[0033] Preferably, the first area point cloud data also includes point cloud height;

[0034] The step of obtaining point cloud change data of each point cloud in a preset period based on the point cloud data of the first area specifically includes:

[0035] Based on the point cloud height, obtaining point cloud height change data of each point cloud in a preset period;

[0036] The step of calculating the difference between the current point cloud quantity and the initial point cloud quantity specifically includes:

[0037] When the point cloud height change data reaches a risk threshold, a point cloud quantity difference between the current point cloud quantity and the initial point cloud quantity is calculated.

[0038] The present invention also provides a pot overflow detection system, the pot overflow detection system comprising: a first area point cloud data acquisition module, a point cloud change data calculation module and a pot overflow control signal generation module;

[0039] The first area point cloud data acquisition module is used to acquire the first area point cloud data within a first preset distance from the pot edge in real time;

[0040] The point cloud change data calculation module is used to obtain the point cloud change data of each point cloud in a preset period based on the point cloud data of the first area;

[0041] The overflow control signal generating module is used to generate an overflow control signal to turn off the fire source of the stove if the point cloud change data reaches the overflow threshold.

[0042] Preferably, the overflow detection system further comprises: a total area point cloud data acquisition module, a grayscale image calculation module and a pot cover position calculation module;

[0043] The total area point cloud data acquisition module is used to acquire the total area point cloud data of the total detection area, and the total detection area covers the stove surface;

[0044] The grayscale image calculation module is used to obtain a grayscale image containing the pot based on the total area point cloud data;

[0045] The pot cover position calculation module is used to identify the grayscale image to obtain the position of the pot cover;

[0046] The first area point cloud data acquisition module is specifically used to extract the first area point cloud data from the total area point cloud data according to the position of the pot cover.

[0047] Preferably, the pot overflow detection system further comprises: a second area point cloud data acquisition module, a plane map calculation module and a pot edge position determination module;

[0048] The second area point cloud data acquisition module is used to extract the second area point cloud data within a second preset distance from the edge of the pot cover from the total area point cloud data according to the position of the pot cover;

[0049] The plan view calculation module is used to obtain a plan view including the cookware based on the second area point cloud data;

[0050] The pot edge position determination module is used to determine the edge position of the pot by extracting arcs based on the plan view;

[0051] The first area point cloud data acquisition module is specifically used to extract the first area point cloud data from the total area point cloud data according to the edge position.

[0052] Preferably, the first area point cloud data acquisition module is specifically used to extract point cloud data within a first preset distance range from the edge position from the total area point cloud data based on a plane fast search algorithm to obtain the first area point cloud data.

[0053] Preferably, the total area point cloud data acquisition module includes: a shooting coordinate point cloud data unit and a coordinate conversion unit;

[0054] The first point cloud data unit is used to obtain point cloud data of the total detection area in the reference coordinate system of the shooting device;

[0055] The coordinate conversion unit is used to obtain the total area point cloud data in the geodetic coordinate system according to the conversion relationship between the shooting device reference coordinate system and the geodetic coordinate system.

[0056] Preferably, the first area point cloud data includes point cloud quantity, and the overflow detection system further includes: an initial point cloud quantity acquisition module;

[0057] The initial point cloud quantity acquisition module is used to acquire the initial point cloud quantity corresponding to the initial time point;

[0058] The overflow control signal generating module is specifically used to calculate the point cloud quantity difference between the current point cloud quantity and the initial point cloud quantity. If the point cloud quantity difference reaches the overflow threshold, an overflow control signal is generated to turn off the fire source of the stove.

[0059] Preferably, the first area point cloud data also includes point cloud height;

[0060] The point cloud change data calculation module is specifically used to obtain the point cloud height change data of each point cloud in a preset period based on the point cloud height;

[0061] The overflow control signal generating module is specifically used to calculate the point cloud quantity difference between the current point cloud quantity and the initial point cloud quantity when the point cloud height change data reaches a risk threshold.

[0062] The present invention also provides a range hood, comprising the aforementioned pan overflow detection system.

[0063] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and the computer program implements the aforementioned overflow detection method when executed by a processor.

[0064] The positive and progressive effect of the present invention is that: through the point cloud data of the first area within the first preset distance range from the edge of the pot, the point cloud change data of each point cloud in the preset period is obtained. If the point cloud change data reaches the overflow threshold, an overflow control signal is generated to turn off the stove fire source, that is, whether there is an overflow is judged by the change of the point cloud at the edge of the pot, and the stove fire source is turned off when the pot overflows. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a flow chart of the overflow detection method of Example 1 of the present invention.

[0066] Figure 2 This is a flow chart of the first specific implementation of the overflow detection method of Example 1 of the present invention.

[0067] Figure 3 This is a flow chart of a second specific implementation of the overflow detection method of Example 1 of the present invention.

[0068] Figure 4 This is a flow chart of the third specific implementation manner of the overflow detection method of Example 1 of the present invention.

[0069] Figure 5 This is a flow chart of the fourth specific implementation of the overflow detection method of Example 1 of the present invention.

[0070] Figure 6 This is a flow chart of the fifth specific implementation method of the overflow detection method in Example 1 of the present invention.

[0071] Figure 7 This is a module schematic diagram of the overflow detection system of Example 2 of the present invention. DETAILED DESCRIPTION

[0072] The present invention is further described below by way of examples, but the present invention is not limited to the scope of the examples.

[0073] Example 1

[0074] This embodiment provides a method for detecting an overflowing pan. Figure 1 , overflow detection methods include:

[0075] S11 , acquiring point cloud data of a first area within a first preset distance range from an edge of a pot in real time.

[0076] S12. Obtaining point cloud change data of each point cloud in a preset period based on the point cloud data of the first region.

[0077] S13: If the point cloud change data reaches the pot overflow threshold, an overflow control signal is generated to turn off the stove fire source.

[0078] Among them, the point cloud data can be collected by the TOF camera (shooting equipment), the transmitting end of the TOF camera reflects infrared light, and the receiving end of the TOF camera receives the reflected infrared light. The TOF camera can be installed on the lower side of the range hood, such as the lower side of the start switch or the lower side of the range hood. The lens of the TOF camera is at a certain angle to the horizontal plane, so that the total detection area of ​​the TOF camera can cover the entire stove surface and avoid oil smoke and water vapor from contaminating the lens.

[0079] Overflowing is caused by too much food in the pot and too much fire, which causes the soup to boil violently and overflow. When overflowing occurs, a large amount of foam flows out along the edge of the pot, causing the circular contour of the edge of the pot to change. By identifying the change in the contour of the edge of the pot cover, it can be determined that the current state is overflowing. The first preset distance range from the edge of the pot can be 1 cm from the edge of the pot.

[0080] This embodiment obtains point cloud change data of each point cloud in a preset period through the point cloud data of the first area within a first preset distance range from the edge of the pot. If the point cloud change data reaches the overflow threshold, an overflow control signal is generated to turn off the stove fire source, that is, whether there is overflow is judged by the change of the point cloud at the edge of the pot, and the stove fire source is turned off when the pot overflows.

[0081] This embodiment provides a first specific implementation method of the overflow detection method. Figure 2 Before step S11, the overflow detection method further includes:

[0082] S101, obtaining total area point cloud data of a total detection area, where the total detection area covers a stove surface.

[0083] S102: Obtain a grayscale image containing the cookware based on the total area point cloud data.

[0084] S103, identifying the grayscale image to obtain the position of the pot cover.

[0085] Step S11 specifically includes:

[0086] S111 . Extracting first area point cloud data from the total area point cloud data according to the position of the pot cover.

[0087] Among them, the three-dimensional point cloud data is projected onto a two-dimensional plane to obtain the grayscale image of the current stove surface as the input parameter of the object recognition module. The object recognition module is built in advance, and the grayscale images of various pots placed on the stove scene collected offline are used as the training set. Through the convolutional neural network training, it can accurately identify the pot cover in the grayscale image. According to the pot cover identified by the object recognition module, the position of the pot cover in the picture is obtained, and then the (x, y) coordinates of the pot cover in the stove scene are calculated.

[0088] This embodiment converts the three-dimensional point cloud data into a two-dimensional grayscale image, and can use a trained object recognition module to accurately identify the pot cover in the grayscale image, thereby obtaining the position of the pot cover, and extracting the first area point cloud data according to the position of the pot cover.

[0089] This embodiment provides a second specific implementation of the overflow detection method based on the first specific implementation. Figure 3 , step S111 specifically includes:

[0090] S1111. Extract second area point cloud data within a second preset distance range from the edge of the pot cover from the total area point cloud data according to the position of the pot cover.

[0091] S1112. Obtain a plan view including the cookware based on the second area point cloud data.

[0092] S1113. Determine the edge position of the pot by extracting arcs based on the plan view.

[0093] S1114. Extract first area point cloud data from the total area point cloud data according to the edge position.

[0094] The second preset distance range from the edge of the pot cover may be 5 cm above and below the edge of the pot cover. The purpose of extracting the second area point cloud data is to obtain a complete pot cover point cloud and remove other point cloud data such as the stove surface.

[0095] Based on the plane diagram, the circular arc contour of the pot is extracted to determine the edge position of the pot. The edge of the pot presents an ellipse (a circle is a special case of an ellipse) shape on the plane. The equation of the ellipse can be expressed as Ax 2 +Bxy+Cy 2 +Dx+Ey+F=0. According to the principle of least squares, the pixel coordinates of the pot cover are substituted into an equation that can fit several ellipses (for example, when the pot cover is half-open, there are multiple elliptical contours) to determine the edge position of the pot.

[0096] This embodiment obtains the complete pot cover point cloud by extracting the second area point cloud data according to the position of the pot cover, and removes other point cloud data such as the stove surface, obtains a plan view containing the pot based on the second area point cloud data, extracts the arc contour of the pot based on the plan view, determines the edge position of the pot, and extracts the first area point cloud data based on the edge position.

[0097] In specific implementation, step S1114 specifically includes:

[0098] Point cloud data within a first preset distance range from an edge position is extracted from the total area point cloud data based on a plane fast search algorithm to obtain first area point cloud data.

[0099] Among them, the plane fast search algorithm includes the kd tree algorithm (an algorithm that uses a data structure that partitions the k-dimensional data space), the binary tree algorithm, the balanced binary tree algorithm, and the like.

[0100] This embodiment provides a specific implementation algorithm for extracting point cloud data of the first area.

[0101] This embodiment provides a third specific implementation of the overflow detection method based on the first specific implementation. Figure 4 , step S101 specifically includes:

[0102] S1011. Obtain point cloud data of the total detection area in the reference coordinate system of the shooting device.

[0103] S1012. Obtain total area point cloud data in the geodetic coordinate system according to a conversion relationship between the photographing device reference coordinate system and the geodetic coordinate system.

[0104] Among them, since the lens of the TOF camera (shooting device) is at a certain angle to the horizontal plane, the point cloud data obtained by the TOF camera is the point cloud data in the reference coordinate system of the shooting device, and the total area point cloud data in the geodetic coordinate system needs to be obtained through coordinate conversion. After the coordinate conversion, the total area point cloud data can be equivalent to that obtained by shooting with a TOF camera with a horizontally placed lens.

[0105] This embodiment converts the point cloud data of the total detection area in the reference coordinate system of the shooting device into the point cloud data of the total area in the geodetic coordinate system through coordinate transformation, optimizes the observation angle of the point cloud data, makes the point cloud changes more intuitive, and is more conducive to subsequent overflow judgment.

[0106] This embodiment provides a fourth specific implementation method of the overflow detection method, referring to Figure 5 The first area point cloud data includes the number of point clouds, and the overflow detection method further includes:

[0107] S120, obtaining the number of initial point clouds corresponding to the initial time point.

[0108] The order of steps S120 and S12 can be interchanged.

[0109] Step S13 specifically includes:

[0110] S131. Calculate the difference between the current point cloud quantity and the initial point cloud quantity. If the difference reaches the pot overflow threshold, generate an overflow control signal to turn off the stove fire source.

[0111] When a normal pot is projected onto a plane, its boundary is clear and there is no noise. The result of an overflowing pot is that a large amount of foam flows out along the edge of the pot. Due to the large amount of foam, there are many noise points at the edge of the pot. By comparing the number of point clouds at the edge of the pot with the number of initial point clouds, it is possible to determine whether a lot of noise points have been added, and thus whether there is an overflowing pot.

[0112] This embodiment can determine whether a lot of noise points are added to the edge of the pot by comparing the number of point clouds in the first area point cloud data with the number of initial point clouds, thereby determining whether there is pot overflow.

[0113] This embodiment provides a fifth specific implementation of the overflow detection method based on the fourth specific implementation. Figure 6 , the first area point cloud data also includes point cloud height.

[0114] Step S12 specifically includes:

[0115] S121. Obtaining point cloud height change data of each point cloud in a preset period based on the point cloud height.

[0116] Step S131 specifically includes:

[0117] S1311. When the point cloud height change data reaches the risk threshold, the point cloud quantity difference between the current point cloud quantity and the initial point cloud quantity is calculated. If the point cloud quantity difference reaches the overflow threshold, an overflow control signal is generated to turn off the stove fire source.

[0118] Among them, when the pot overflows, a large amount of foam will both be generated and burst, and the generation and bursting of foam will cause the height value measured at the edge of the pot to change.

[0119] In this embodiment, when the point cloud height change data reaches the risk threshold, that is, when the point cloud height change data reflects that a certain risk of pot overflow has been reached, it is determined whether a lot of noise points are added to the edge of the pot, thereby determining whether there is pot overflow.

[0120] By converting three-dimensional point cloud data into a two-dimensional grayscale image, this embodiment can use a trained object recognition module to accurately identify the pot cover in the grayscale image, and then obtain the position of the pot cover, extract the second area point cloud data according to the position of the pot cover to obtain a complete pot cover point cloud, and remove other point cloud data such as the stove surface, obtain a plane view containing the pot based on the second area point cloud data, extract the arc contour of the pot based on the plane view, determine the edge position of the pot, extract the first area point cloud data according to the edge position, and obtain point cloud change data of each point cloud in a preset period based on the first area point cloud data. If the point cloud change data reaches the overflow threshold, an overflow control signal is generated to turn off the stove fire source, that is, whether there is overflow is determined by the change of the point cloud at the edge of the pot, and the stove fire source is turned off when the pot overflows. This embodiment also converts the point cloud data of the total detection area in the reference coordinate system of the shooting device into the point cloud data of the total area in the geodetic coordinate system through coordinate transformation, optimizes the observation angle of the point cloud data, makes the point cloud changes more intuitive, and is more conducive to subsequent pot overflow judgment. When the point cloud height change data reaches the risk threshold, that is, when the point cloud height change data reflects that a certain pot overflow risk has been reached, it is judged whether a lot of noise points are added to the edge of the pot, thereby judging whether there is a pot overflow.

[0121] Example 2

[0122] This embodiment provides a pan overflow detection system. Figure 7 The overflow detection system includes: a first area point cloud data acquisition module 1, a point cloud change data calculation module 2 and an overflow control signal generation module 3.

[0123] The first area point cloud data acquisition module 1 is used to acquire the first area point cloud data within a first preset distance range from the pot edge in real time.

[0124] The point cloud change data calculation module 2 is used to obtain the point cloud change data of each point cloud in a preset period based on the point cloud data of the first area.

[0125] The overflow control signal generating module 3 is used to generate an overflow control signal to turn off the fire source of the stove if the point cloud change data reaches the overflow threshold.

[0126] Among them, the point cloud data can be collected by the TOF camera (shooting equipment), the transmitting end of the TOF camera reflects infrared light, and the receiving end of the TOF camera receives the reflected infrared light. The TOF camera can be installed on the lower side of the range hood, such as the lower side of the start switch or the lower side of the range hood. The lens of the TOF camera is at a certain angle to the horizontal plane, so that the total detection area of ​​the TOF camera can cover the entire stove surface and avoid oil smoke and water vapor from contaminating the lens.

[0127] Overflowing is caused by too much food in the pot and too much fire, which causes the soup to boil violently and overflow. When overflowing occurs, a large amount of foam flows out along the edge of the pot, causing the circular contour of the edge of the pot to change. By identifying the change in the contour of the edge of the pot cover, it can be determined that the current state is overflowing. The first preset distance range from the edge of the pot can be 1 cm from the edge of the pot.

[0128] This embodiment obtains point cloud change data of each point cloud in a preset period through the point cloud data of the first area within a first preset distance range from the edge of the pot. If the point cloud change data reaches the overflow threshold, an overflow control signal is generated to turn off the stove fire source, that is, whether there is overflow is judged by the change of the point cloud at the edge of the pot, and the stove fire source is turned off when the pot overflows.

[0129] In specific implementation, the overflow detection system also includes: a total area point cloud data acquisition module 4, a grayscale image calculation module 5 and a pot cover position calculation module 6.

[0130] The total area point cloud data acquisition module 4 is used to acquire the total area point cloud data of the total detection area, and the total detection area covers the stove surface.

[0131] The grayscale image calculation module 5 is used to obtain a grayscale image containing the pot based on the total area point cloud data.

[0132] The pot cover position calculation module 6 is used to identify the grayscale image to obtain the position of the pot cover.

[0133] The first area point cloud data acquisition module 1 is specifically used to extract the first area point cloud data from the total area point cloud data according to the position of the pot cover.

[0134] Among them, the three-dimensional point cloud data is projected onto a two-dimensional plane to obtain the grayscale image of the current stove surface as the input parameter of the object recognition module. The object recognition module is built in advance, and the grayscale images of various pots placed on the stove scene collected offline are used as the training set. Through the convolutional neural network training, it can accurately identify the pot cover in the grayscale image. According to the pot cover identified by the object recognition module, the position of the pot cover in the picture is obtained, and then the (x, y) coordinates of the pot cover in the stove scene are calculated.

[0135] This embodiment converts the three-dimensional point cloud data into a two-dimensional grayscale image, and can use a trained object recognition module to accurately identify the pot cover in the grayscale image, thereby obtaining the position of the pot cover, and extracting the first area point cloud data according to the position of the pot cover.

[0136] In specific implementation, the pot overflow detection system further includes: a second area point cloud data acquisition module 7, a plane map calculation module 8 and a pot edge position determination module 9.

[0137] The second area point cloud data acquisition module 7 is used to extract the second area point cloud data within a second preset distance range from the edge of the pot cover from the total area point cloud data according to the position of the pot cover.

[0138] The plan view calculation module 8 is used to obtain a plan view including the cookware based on the second area point cloud data.

[0139] The pot edge position determination module 9 is used to determine the pot edge position by extracting arcs based on the plan view.

[0140] The first area point cloud data acquisition module 1 is specifically used to extract the first area point cloud data from the total area point cloud data according to the edge position.

[0141] The second preset distance range from the edge of the pot cover may be 5 cm above and below the edge of the pot cover. The purpose of extracting the second area point cloud data is to obtain a complete pot cover point cloud and remove other point cloud data such as the stove surface.

[0142] Based on the plane diagram, the arc contour of the pot is extracted to determine the edge position of the pot. The edge of the pot presents an ellipse (a circle is a special case of an ellipse) shape on the plane. The equation of the ellipse can be expressed as Ax 2 +Bxy+Cy 2 +Dx+Ey+F=0. According to the principle of least squares, the pixel coordinates of the pot cover are substituted into an equation that can fit several ellipses (for example, when the pot cover is half-open, there are multiple elliptical contours) to determine the edge position of the pot.

[0143] This embodiment obtains the complete pot cover point cloud by extracting the second area point cloud data according to the position of the pot cover, and removes other point cloud data such as the stove surface, obtains a plan view containing the pot based on the second area point cloud data, extracts the arc contour of the pot based on the plan view, determines the edge position of the pot, and extracts the first area point cloud data based on the edge position.

[0144] In specific implementation, the first area point cloud data acquisition module 1 is specifically used to extract point cloud data within a first preset distance range from the edge position from the total area point cloud data based on a plane fast search algorithm to obtain the first area point cloud data.

[0145] Among them, the plane fast search algorithm includes the kd tree algorithm, the binary tree algorithm, the balanced binary tree algorithm, etc.

[0146] This embodiment provides a specific implementation algorithm for extracting point cloud data of the first area.

[0147] In specific implementation, the total area point cloud data acquisition module 4 includes: a shooting coordinate point cloud data unit 401 and a coordinate conversion unit 402 .

[0148] The shooting coordinate point cloud data unit 401 is used to obtain the point cloud data of the total detection area in the reference coordinate system of the shooting device.

[0149] The coordinate conversion unit 402 is used to obtain the total area point cloud data in the geodetic coordinate system according to the conversion relationship between the photographing device reference coordinate system and the geodetic coordinate system.

[0150] Among them, since the lens of the TOF camera (shooting device) is at a certain angle to the horizontal plane, the point cloud data obtained by the TOF camera is the point cloud data in the reference coordinate system of the shooting device, and the total area point cloud data in the geodetic coordinate system needs to be obtained through coordinate conversion. After the coordinate conversion, the total area point cloud data can be equivalent to that obtained by shooting with a TOF camera with a horizontally placed lens.

[0151] This embodiment converts the point cloud data of the total detection area in the reference coordinate system of the shooting device into the point cloud data of the total area in the geodetic coordinate system through coordinate transformation, optimizes the observation angle of the point cloud data, makes the point cloud changes more intuitive, and is more conducive to subsequent overflow judgment.

[0152] During specific implementation, the first area point cloud data includes the number of point clouds, and the overflow detection system further includes: an initial point cloud number acquisition module 10.

[0153] The initial point cloud quantity acquisition module 10 is used to acquire the initial point cloud quantity corresponding to the initial time point.

[0154] The overflow control signal generating module 3 is specifically used to calculate the point cloud quantity difference between the current point cloud quantity and the initial point cloud quantity. If the point cloud quantity difference reaches the overflow threshold, an overflow control signal is generated to turn off the stove fire source.

[0155] When a normal pot is projected onto a plane, its boundary is clear and there is no noise. The result of an overflowing pot is that a large amount of foam flows out along the edge of the pot. Due to the large amount of foam, there are many noise points at the edge of the pot. By comparing the number of point clouds at the edge of the pot with the number of initial point clouds, it is possible to determine whether a lot of noise points have been added, and thus whether there is an overflowing pot.

[0156] This embodiment can determine whether a lot of noise points are added to the edge of the pot by comparing the number of point clouds in the first area point cloud data with the number of initial point clouds, thereby determining whether there is pot overflow.

[0157] During specific implementation, the first area point cloud data also includes point cloud height.

[0158] The point cloud change data calculation module 2 is specifically used to obtain the point cloud height change data of each point cloud in a preset period based on the point cloud height.

[0159] The overflow control signal generating module 3 is specifically used to calculate the point cloud quantity difference between the current point cloud quantity and the initial point cloud quantity when the point cloud height change data reaches the risk threshold.

[0160] Among them, when the pot overflows, a large amount of foam will both be generated and burst, and the generation and bursting of foam will cause the height value measured at the edge of the pot to change.

[0161] In this embodiment, when the point cloud height change data reaches the risk threshold, that is, when the point cloud height change data reflects that a certain risk of pot overflow has been reached, it is determined whether a lot of noise points are added to the edge of the pot, thereby determining whether there is pot overflow.

[0162] By converting three-dimensional point cloud data into a two-dimensional grayscale image, this embodiment can use a trained object recognition module to accurately identify the pot cover in the grayscale image, and then obtain the position of the pot cover, extract the second area point cloud data according to the position of the pot cover to obtain a complete pot cover point cloud, and remove other point cloud data such as the stove surface, obtain a plane view containing the pot based on the second area point cloud data, extract the arc contour of the pot based on the plane view, determine the edge position of the pot, extract the first area point cloud data according to the edge position, and obtain point cloud change data of each point cloud in a preset period based on the first area point cloud data. If the point cloud change data reaches the overflow threshold, an overflow control signal is generated to turn off the stove fire source, that is, whether there is overflow is determined by the change of the point cloud at the edge of the pot, and the stove fire source is turned off when the pot overflows. This embodiment also converts the point cloud data of the total detection area in the reference coordinate system of the shooting device into the point cloud data of the total area in the geodetic coordinate system through coordinate transformation, optimizes the observation angle of the point cloud data, makes the point cloud changes more intuitive, and is more conducive to subsequent pot overflow judgment. When the point cloud height change data reaches the risk threshold, that is, when the point cloud height change data reflects that a certain pot overflow risk has been reached, it is judged whether a lot of noise points are added to the edge of the pot, thereby judging whether there is a pot overflow.

[0163] Example 3

[0164] This embodiment provides a range hood, including the pan overflow detection system in Embodiment 2.

[0165] By converting three-dimensional point cloud data into a two-dimensional grayscale image, this embodiment can use a trained object recognition module to accurately identify the pot cover in the grayscale image, and then obtain the position of the pot cover, extract the second area point cloud data according to the position of the pot cover to obtain a complete pot cover point cloud, and remove other point cloud data such as the stove surface, obtain a plane view containing the pot based on the second area point cloud data, extract the arc contour of the pot based on the plane view, determine the edge position of the pot, extract the first area point cloud data according to the edge position, and obtain point cloud change data of each point cloud in a preset period based on the first area point cloud data. If the point cloud change data reaches the overflow threshold, an overflow control signal is generated to turn off the stove fire source, that is, whether there is overflow is determined by the change of the point cloud at the edge of the pot, and the stove fire source is turned off when the pot overflows. This embodiment also converts the point cloud data of the total detection area in the reference coordinate system of the shooting device into the point cloud data of the total area in the geodetic coordinate system through coordinate transformation, optimizes the observation angle of the point cloud data, makes the point cloud changes more intuitive, and is more conducive to subsequent pot overflow judgment. When the point cloud height change data reaches the risk threshold, that is, when the point cloud height change data reflects that a certain pot overflow risk has been reached, it is judged whether a lot of noise points are added to the edge of the pot, thereby judging whether there is a pot overflow.

[0166] Example 4

[0167] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the overflow detection method provided in Embodiment 1 is implemented.

[0168] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0169] In a possible implementation manner, the present invention can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the overflow detection method described in Example 1.

[0170] The program code for executing the present invention may be written in any combination of one or more programming languages, and may be executed entirely on a user device, partially on a user device, as an independent software package, partially on a user device and partially on a remote device, or entirely on a remote device.

[0171] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that this is only for illustration and the protection scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.

Claims

1. A method for detecting an overflowing pan, characterized in that: The overflow detection method comprises: Acquire point cloud data of a first area within a first preset distance range from the edge of the pot in real time; Obtaining point cloud change data of each point cloud in a preset period based on the point cloud data of the first area; If the point cloud change data reaches the pot overflow threshold, an overflow control signal is generated to turn off the stove fire source; Before the step of acquiring point cloud data of a first area within a preset distance range from the pot edge in real time, the pot overflow detection method further includes: Acquire total area point cloud data of a total detection area, where the total detection area covers a stove surface; Obtaining a grayscale image containing the pot based on the total area point cloud data; Identify the grayscale image to obtain the position of the pot cover; The step of acquiring point cloud data of a first area within a first preset distance from the edge of the pot in real time specifically comprises: Extracting second area point cloud data within a second preset distance from the edge of the pot cover from the total area point cloud data according to the position of the pot cover; Obtaining a plan view including the pot based on the second area point cloud data; Determine the edge position of the pot by extracting arcs based on the plan view; The point cloud data of the first area is obtained by extracting point cloud data within a first preset distance range from the edge position from the point cloud data of the total area based on a plane fast search algorithm.

2. The overflow detection method according to claim 1, characterized in that: The step of obtaining the total area point cloud data of the total detection area specifically includes: Obtain point cloud data of the total detection area in the reference coordinate system of the shooting device; The total area point cloud data in the geodetic coordinate system is obtained according to the conversion relationship between the shooting device reference coordinate system and the geodetic coordinate system.

3. The overflow detection method according to claim 1, characterized in that: The first area point cloud data includes the number of point clouds, and the overflow detection method further includes: Get the number of initial point clouds corresponding to the initial time point; If the point cloud change data reaches the pot overflow threshold, the step of generating a pot overflow control signal to turn off the stove fire source specifically includes: The point cloud quantity difference between the current point cloud quantity and the initial point cloud quantity is calculated, and if the point cloud quantity difference reaches the pot overflow threshold, an overflow control signal is generated to turn off the fire source of the stove.

4. The overflow detection method according to claim 3, characterized in that: The first area point cloud data also includes point cloud height; The step of obtaining point cloud change data of each point cloud in a preset period based on the point cloud data of the first area specifically includes: Based on the point cloud height, obtaining point cloud height change data of each point cloud in a preset period; The step of calculating the difference between the current point cloud quantity and the initial point cloud quantity specifically includes: When the point cloud height change data reaches a risk threshold, a point cloud quantity difference between the current point cloud quantity and the initial point cloud quantity is calculated.

5. An overflow detection system, characterized in that: The overflow detection system comprises: a first area point cloud data acquisition module, a point cloud change data calculation module and an overflow control signal generation module; The first area point cloud data acquisition module is used to acquire the first area point cloud data within a first preset distance from the pot edge in real time; The point cloud change data calculation module is used to obtain the point cloud change data of each point cloud in a preset period based on the point cloud data of the first area; The overflow control signal generating module is used for generating an overflow control signal to turn off the fire source of the stove if the point cloud change data reaches the overflow threshold; The overflow detection system further includes: a total area point cloud data acquisition module, a grayscale image calculation module and a pot cover position calculation module; The total area point cloud data acquisition module is used to acquire the total area point cloud data of the total detection area, and the total detection area covers the stove surface; The grayscale image calculation module is used to obtain a grayscale image containing the pot based on the total area point cloud data; The pot cover position calculation module is used to identify the grayscale image to obtain the position of the pot cover; The overflow pot detection system further comprises: a second area point cloud data acquisition module, a plane map calculation module and a pot edge position determination module; The second area point cloud data acquisition module is used to extract the second area point cloud data within a second preset distance from the edge of the pot cover from the total area point cloud data according to the position of the pot cover; The plan view calculation module is used to obtain a plan view including the cookware based on the second area point cloud data; The pot edge position determination module is used to determine the edge position of the pot by extracting arcs based on the plan view; The first area point cloud data acquisition module is specifically used to extract point cloud data within a first preset distance range from the edge position from the total area point cloud data based on a plane fast search algorithm to obtain the first area point cloud data.

6. A range hood, characterized in that: Includes the overflow detection system as described in claim 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the overflow detection method according to any one of claims 1 to 4 is implemented.

Images