Hot pot induction cooker fire control method based on vision

Abstract

The invention discloses a hot pot induction cooker firepower control method based on vision, and the method comprises the following steps: collecting table region image data through a camera, and transmitting the data to a data processing part through a network, a high-performance computer calculates and analyzes image data acquired from a camera to obtain an analysis result, firepower is controlled through a network according to the analysis result, and an intelligent induction cooker is in communication connection with the computer, receives a control signal to adjust the firepower and receives a feedback signal. The feedback signal receiving is mainly characterized in that after the firepower of the intelligent induction cooker is adjusted, the boiling condition of the bottom of the hot pot is continuously collected through cameras, the cameras in the S1 are in communication connection with a computer through a network, and the multiple cameras are installed above the dining area of the hot pot. According to the invention, human intervention actions can be reduced, energy consumption is saved, and the dining experience of customers is improved.

Description

Technical Field

[0001] The invention relates to the technical field of hot pot induction cooker fire control, and in particular to a vision-based hot pot induction cooker fire control method. Background Art

[0002] At present, hot pot is mainly heated by induction cookers. In reality, most hot pot restaurants have problems such as the ingredients at the bottom of the pot boiling for a long time after being cooked while customers are eating, and the bottom of the pot continues to boil after customers stop eating. This will cause a lot of electricity waste. The current solution to this problem mainly relies on manual observation by service personnel and adjustment of the firepower according to the on-site situation.

[0003] However, manual adjustment may disturb customers and affect their dining experience. At the same time, there is a large flow of people during peak hours, and service staff cannot adjust or turn off the hot pot that has been boiling for a long time in time. Therefore, there is an urgent need for a visual-based hot pot induction cooker fire control method that can adjust the fire of the induction cooker based on the boiling condition of the pot bottom and the customer's dining situation. It can reduce part of the firepower when the boiling degree is high, increase part of the firepower when the customer's dining frequency is high, and reduce or turn off the firepower when the customer is not dining. Summary of the invention

[0004] The purpose of the present invention is to provide a vision-based method for controlling the firepower of a hot pot induction cooker to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a vision-based hot pot induction cooker firepower control method, the control method comprising the following steps:

[0006] S1, collect table area image data through the camera and send it to the data processing part through the network;

[0007] S2, the data processing part is composed of a computer, and the image data obtained from the camera is analyzed by a high-performance computer to obtain an analysis result;

[0008] S3, according to the analysis results, the firepower is controlled through the network, the intelligent induction cooker is connected to the computer through communication, and the intelligent induction cooker receives the control signal to adjust the firepower and receive the feedback signal;

[0009] S4. The receiving feedback signal is mainly to continuously collect the boiling condition of the hot pot bottom through the camera after the firepower of the intelligent induction cooker is adjusted.

[0010] Preferably, the camera in S1 is connected to the computer through a network, and multiple cameras are installed above the hot pot dining area to collect image data of the boiling conditions of the hot pot bottom and dining behavior, and the collected image data is sent to the computer through the network.

[0011] Preferably, the main steps of obtaining the analysis result by analyzing the image data acquired from the camera through high-performance computer calculation include:

[0012] 1) Obtain the current image data through the network camera, and cut the acquired image data to obtain the dining area image Roi t ;

[0013] 2) At the same time, the dining area image Roi t Cut to get the pot bottom area image PotRoi t ;

[0014] 3) The dining area image Roi t PotRoi with pot bottom area image t Resize to 224×224 and convert the dining area image Roi t Store in a queue Image with a maximum length of 16;

[0015] 4) Use the image classification algorithm to classify the pot bottom area image and obtain the classification result c t

[0016] c t =Classification(PotRoi t );

[0017] 5) Store the classification results in a queue C with a maximum length of 16;

[0018] 6) When t≥16, the image queue Image={Roi1,Roi2,…,Roi 16}, the pot bottom classification queue C = {c1,c2,…,c 16}. When t<16, repeat steps 1) to 5);

[0019] 7) Send the obtained image queue to the spatiotemporal action detection network Yowo to obtain the detection result y t .

[0020] y t =Yowo(Image);

[0021] 8) Determine whether there is a "dining" behavior in the current detection result classification. If so, store 1 in the queue Y with a maximum length of 20. If not, store 0 in the queue Y.

[0022] 9) At the same time, calculate the number of each category in the pot bottom classification queue C, and store the categories with a number of categories greater than 8 in the queue B with a maximum length of 20;

[0023] 10) Count the number of identical elements in queue Y and queue B. If the number of 1s in Y is greater than 10, that is, the number of dining behaviors is greater than 10 and less than 15, and the number of boilings in B is greater than 10, then the firepower of the smart induction cooker is reduced by one level; if the number of 1s in Y is greater than 15 and the number of boilings in B is greater than 10, then the firepower of the smart induction cooker is not adjusted; if the number of 1s in Y is less than 10 and the number of boilings in B is greater than 10, then the firepower of the smart induction cooker is reduced by two levels; if the number of 1s in Y is less than 10 and the number of boilings in B is less than 10, then the firepower of the smart induction cooker is not adjusted; if the number of 1s in Y is less than 5 and the number of boilings in B is less than 10, then the firepower of the induction cooker is adjusted to the minimum; if the number of boilings in B is less than 5, then the firepower of the smart induction cooker is not adjusted.

[0024] 1. Based on the image data sampled from store surveillance, the surveillance utilization rate is improved without any additional cost;

[0025] 2. Use image processing technology to analyze image information, reduce manual intervention, and improve dining experience and service quality;

[0026] 3. Basically monitor and respond to the status of the pot bottom in real time, with a delay of less than 15s;

[0027] 3. Process the results through statistical methods to ensure the stability and robustness of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A flow chart of a vision-based hot pot induction cooker fire control method Figure 1 ;

[0029] Figure 2 A flow chart of a vision-based hot pot induction cooker fire control method Figure 2 ;

[0030] Figure 3 A flow chart of a vision-based hot pot induction cooker fire control method Figure 2 . DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] See also Figure 1-3 The present invention provides a technical solution: a vision-based hot pot induction cooker fire control method, the control method comprising the following steps:

[0033] S1. Collect the image data of the table area through the camera and send it to the data processing part through the network. Access the camera's dedicated app through the network to transmit the collected data in real time. The camera should preferably have an image of more than 380,000 dots, a color resolution greater than or equal to 480 lines, and a black and white resolution of more than 600 lines;

[0034] S2, the data processing part is composed of a computer, which uses high-performance computers to calculate and analyze the image data obtained from the camera to obtain the analysis results;

[0035] S3, according to the analysis results, the firepower is controlled through the network, the intelligent induction cooker is connected to the computer through communication, and the intelligent induction cooker receives the control signal to adjust the firepower and receive the feedback signal;

[0036] S4, receiving feedback signals mainly involves continuously collecting the boiling conditions of the hot pot bottom through a camera after the firepower of the smart induction cooker is adjusted.

[0037] The camera in S1 is connected to the computer through a network. Multiple cameras are installed above the hot pot dining area to collect image data of the boiling conditions of the hot pot bottom and dining behavior, which includes the action of picking up food and eating. The collected image data is sent to the computer through the network.

[0038] The main steps of obtaining the analysis results by analyzing the image data obtained from the camera through high-performance computer calculations include:

[0039] 1) Obtain the current image data through the network camera, and cut the acquired image data to obtain the dining area image Roi t ;

[0040] 2) At the same time, the dining area image Roi t Cut to get the pot bottom area image PotRoi t ;

[0041] 3) The dining area image Roi t PotRoi with pot bottom area imaget Resize to 224×224 and convert the dining area image Roi t Store in a queue Image with a maximum length of 16;

[0042] 4) Use the image classification algorithm to classify the pot bottom area image and obtain the classification result c t

[0043] c t =Classification(PotRoi t );

[0044] 5) Store the classification results in a queue C with a maximum length of 16;

[0045] 6) When t≥16, the image queue Image={Roi1,Roi2,…,Roi 16}, the pot bottom classification queue C = {c1,c2,…,c 16}. When t<16, repeat steps 1) to 5);

[0046] 7) The resulting image queue is fed into the spatiotemporal action detection network Yowo. YOWO is a single-stage architecture with two branches that can extract temporal and spatial information simultaneously in one evaluation and directly predict bounding boxes and action probabilities from video clips to obtain the detection result y t .

[0047] y t =Yowo(Image);

[0048] 8) Determine whether there is a "dining" behavior in the current detection result classification. If so, store 1 in the queue Y with a maximum length of 20. If not, store 0 in the queue Y.

[0049] 9) At the same time, calculate the number of each category in the pot bottom classification queue C, and store the categories with a number of categories greater than 8 in the queue B with a maximum length of 20;

[0050] 10) Count the number of identical elements in queue Y and queue B. If the number of 1s in Y is greater than 10, that is, the number of dining behaviors is greater than 10 and less than 15, and the number of boilings in B is greater than 10, then the firepower of the smart induction cooker is reduced by one gear; if the number of 1s in Y is greater than 15 and the number of boilings in B is greater than 10, then the firepower of the smart induction cooker is not adjusted; if the number of 1s in Y is less than 10 and the number of boilings in B is greater than 10, then the firepower of the smart induction cooker is reduced by two gears; if the number of 1s in Y is less than 10 and the number of boilings in B is less than 10, then the firepower of the smart induction cooker is not adjusted; if the number of 1s in Y is less than 5 and the number of boilings in B is less than 10, then the firepower of the induction cooker is adjusted to the minimum; if the number of boilings in B is less than 5, then the firepower of the smart induction cooker is not adjusted. Please refer to the attached manual for details. Figure 2 .

[0051] Working principle: This method mainly uses cameras to collect the boiling conditions of the hot pot bottom and the dining behavior of people in the dining area. According to whether the people have taken the ingredients on the table or in the pot during their dining behavior, all these data are sent to the background computer for processing and analysis. According to the analysis results, the computer controls the smart induction cooker to increase or reduce the firepower or directly turn off the induction cooker. This method can reduce human intervention, save energy and improve the customer's dining experience.

[0052] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0053] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A vision-based method for controlling the firepower of a hot pot induction cooker, characterized in that: The control method comprises the following steps: S1, collect table area image data through the camera and send it to the data processing part through the network; S2, the data processing part is composed of a computer, and the image data obtained from the camera is analyzed by a high-performance computer to obtain an analysis result; S3, according to the analysis results, the firepower is controlled through the network, the intelligent induction cooker is connected to the computer through communication, and the intelligent induction cooker receives the control signal to adjust the firepower and receive the feedback signal; S4. The receiving feedback signal is mainly to continuously collect the boiling condition of the hot pot bottom through the camera after the firepower of the intelligent induction cooker is adjusted.

2. The vision-based hot pot induction cooker firepower control method according to claim 1, characterized in that: The camera in S1 is connected to the computer through a network. Multiple cameras are installed above the hot pot dining area to collect image data of the boiling conditions of the hot pot bottom and dining behavior, and the collected image data is sent to the computer through the network.

3. The vision-based hot pot induction cooker firepower control method according to claim 1, characterized in that: The main steps of obtaining the analysis result by analyzing the image data obtained from the camera through high-performance computer calculation include: 1) Obtain the current image data through the network camera, and cut the acquired image data to obtain the dining area image Roi t ; 2) At the same time, the dining area image Roi t Cut to get the pot bottom area image PotRoi t ; 3) The dining area image Roi t PotRoi with pot bottom area image t Resize to 224×224 and convert the dining area image Roi t Store in a queue Image with a maximum length of 16; 4) Use the image classification algorithm to classify the pot bottom area image and obtain the classification result c t c t =Classification(PotRoi t ); 5) Store the classification results in a queue C with a maximum length of 16; 6) When t≥16, the image queue Image={Roi1,Roi2,…,Roi 16 }, the pot bottom classification queue C = {c1,c2,…,c 16 }. When t<16, repeat steps 1) to 5); 7) Send the obtained image queue to the spatiotemporal action detection network Yowo to obtain the detection result y t . yes t =Yowo(Image); 8) Determine whether there is a "dining" behavior in the current detection result classification. If so, store 1 in the queue Y with a maximum length of 20. If not, store 0 in the queue Y. 9) At the same time, calculate the number of each category in the pot bottom classification queue C, and store the categories with a number of categories greater than 8 in the queue B with a maximum length of 20; 10) Count the number of identical elements in queue Y and queue B. If the number of 1s in Y is greater than 10, that is, the number of dining behaviors is greater than 10 and less than 15, and the number of boilings in B is greater than 10, then the firepower of the smart induction cooker is reduced by one level; if the number of 1s in Y is greater than 15 and the number of boilings in B is greater than 10, then the firepower of the smart induction cooker is not adjusted; if the number of 1s in Y is less than 10 and the number of boilings in B is greater than 10, then the firepower of the smart induction cooker is reduced by two levels; if the number of 1s in Y is less than 10 and the number of boilings in B is less than 10, then the firepower of the smart induction cooker is not adjusted; if the number of 1s in Y is less than 5 and the number of boilings in B is less than 10, then the firepower of the induction cooker is adjusted to the minimum; if the number of boilings in B is less than 5, then the firepower of the smart induction cooker is not adjusted.

Images