Intelligent range hood device and intelligent control and food recommendation method

Abstract

This solution discloses an intelligent range hood device and an intelligent control and food recommendation method, wherein the first monitoring module is used to monitor the temperature of the target in the pot; the first confirmation module is used to compare the temperature information with the set temperature, and if the temperature information is greater than the set temperature, the second monitoring module is notified to enter the monitoring state; the second monitoring module is used to monitor the oil fume concentration; the second confirmation module is used to compare the oil fume concentration with the first set concentration, and if the oil fume concentration is greater than the first set concentration, the control module is notified to turn on the range hood; the timing is started when the oil fume concentration is less than the second set concentration, and the control module is notified to turn off the range hood after the timing reaches the set time; the control module is used to control the opening and closing of the range hood and the working speed. This solution can avoid indoor environmental pollution or energy waste, and improve the accuracy and stability of intelligent control of the range hood.

Description

Technical Field

[0001] The present invention belongs to the technical field of kitchen appliances, and in particular relates to an intelligent range hood device and an intelligent control and food recommendation method. Background Art

[0002] Range hoods play a very important role in our lives. The existence of range hoods has greatly improved our quality of life and the cleanliness of the air. When people cook, a large amount of exhaust gas is generated, and these exhaust gases are all hot air, which will float above the stove. Placing the range hood above the stove will make it easier for the exhaust gas to enter the range hood. However, the current range hoods are not smart enough. First of all, the range hood needs to be turned on or off manually. In actual use, it is often forgotten to turn on the range hood. If the range hood is not turned on in time during cooking, it will cause oil smoke to enter the home and endanger health. If the range hood is not turned off for a long time after cooking, it will cause a waste of energy. When the range hood is turned off immediately after cooking, it will cause oil smoke to remain, which is also not conducive to health and cleanliness.

[0003] In addition, the existing range hoods still have the problem of single function. As people's living standards improve, smart, environmentally friendly, healthy and diversified food has become a goal that people are increasingly pursuing. Most people in life are often troubled by what to eat, and a diversified diet is loved by the majority of users who pursue health. If the range hood can be used to collect food information, recommend more nutritionally balanced ingredients, and recommend recipes for the corresponding ingredients, it will be more conducive to solving people's daily problems and selecting more nutritionally balanced ingredients. Summary of the invention

[0004] The object of the present invention is to provide an intelligent range hood device and an intelligent control and food recommendation method in view of the above problems.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A range hood intelligent control method comprises the following steps:

[0007] SA1. Monitor the temperature of the target in the pot;

[0008] SA2. When the temperature of the target in the pot exceeds the set temperature, start monitoring the fume concentration;

[0009] SA3. When the oil smoke concentration is greater than the first set concentration, turn on the range hood;

[0010] SA4. Continuously monitor the oil fume concentration, and start timing when the oil fume concentration is less than the second set concentration;

[0011] SA5. Turn off the range hood after the set time is reached.

[0012] In the above-mentioned intelligent control method for range hoods, in step SA4, the oil fume concentration is continuously monitored during the timing period, and if the oil fume concentration is greater than the first set concentration, the timing is terminated, and the timing is restarted until the oil fume concentration is less than the second set concentration again;

[0013] In step AS1, the temperature of the target in the pot is monitored by:

[0014] SA11. Obtain image information of the target in the pot;

[0015] SA12. Use infrared thermal imaging technology to obtain a thermal image based on the image information, thereby obtaining the temperature of the target in the pot;

[0016] In step SA2, the oil fume concentration is monitored by an oil fume sensor, and the oil fume sensor is a particle concentration sensor or a VOC gas sensor;

[0017] In step SA3, the current cooking method is determined according to the recipe selected by the user, and the range hood is controlled to operate at a low speed, a medium speed, or a high speed according to the cooking method.

[0018] A method for recommending food materials for a range hood based on intelligent control of the range hood, comprising the following methods:

[0019] SB1. During the cooking process with the range hood turned on, obtain the image information of the target in the pot, and analyze the image information to determine the food information corresponding to the image information;

[0020] SB2. Convert the food information into the corresponding nutritional element information code for storage;

[0021] SB3. Use the saved historical information codes to recommend ingredients to users; the saved historical information codes include a large amount of information codes saved in several steps SB1 and SB2.

[0022] SB4. Save the recommended ingredients selected by the user, and recommend recipes to the user based on the recommended ingredients selected by the user.

[0023] In the above-mentioned method for recommending ingredients for a range hood based on intelligent control of a range hood, in step SB3, ingredients are recommended to the user by the following method:

[0024] SB31. Analyze the nutritional balance of users based on the user-ingredient preference scoring matrix; the user-ingredient preference scoring matrix is ​​established based on the user's historical information encoding;

[0025] Calculate the similarity between the target user and other users based on the user-ingredient preference score matrix, and obtain the target user's adjacent user group based on the similarity;

[0026] SB33. Select several users whose nutritional balance degree is higher than the balance threshold from the adjacent user group as the neighbor set of the target user;

[0027] SB34. Calculate the preference scores of ingredients for which the target user has not given a preference score or whose preference scores are lower than the preference score threshold, and sort them according to the existing preference scores of the neighbor set;

[0028] SB35. Recommend the N or M ingredients with the highest calculated scores to the target user. When N<M, take N, otherwise take M, where M is the number of ingredients for which ingredient preference scoring was performed in step SB34.

[0029] In the above-mentioned range hood food recommendation method based on intelligent control of the range hood, in step SB1, an infrared thermal imager is used to obtain image information of the target in the pot;

[0030] In step SB4, the recommended recipes are pushed to the user client, and the recommended recipes selected by the user are displayed on the display screen of the range hood.

[0031] An intelligent range hood device comprises a first monitoring module, a first confirmation module, a second monitoring module, a second confirmation module and a control module, wherein:

[0032] A first monitoring module, used for monitoring the temperature of the target in the pot and transmitting the temperature information to a first confirmation module;

[0033] A first confirmation module is used to compare the received temperature information with the set temperature, and if the monitored temperature information is greater than the set temperature, notify the second monitoring module to enter the monitoring state;

[0034] The second monitoring module is used to monitor the oil fume concentration and transmit the oil fume concentration information to the second confirmation module;

[0035] A second confirmation module is used to compare the received oil fume concentration with the first set concentration, and if the monitored oil fume concentration is greater than the first set concentration, notify the control module to turn on the range hood;

[0036] It is also used to continuously monitor the oil smoke concentration when the range hood is turned on, start timing when the oil smoke concentration is less than a second set concentration, and notify the control module to turn off the range hood after the set time is reached;

[0037] The control module is used to control the opening, closing and operating speed of the range hood.

[0038] In the above-mentioned intelligent range hood device, the temperature information transmitted by the first monitoring module to the first confirmation module is the highest temperature of the target in the pot during the current detection;

[0039] Alternatively, the temperature information transmitted by the first monitoring module to the first confirmation module is the average temperature of the target in the pot during the current detection;

[0040] Alternatively, the temperature information transmitted by the first monitoring module to the first confirmation module is the temperature distribution in the current detection, and the first confirmation module selects the highest temperature from the temperature distribution or calculates the average temperature to compare with the set temperature.

[0041] In the above-mentioned intelligent range hood device, the second monitoring module continues to monitor the oil fume concentration during the timing period, and the second confirmation module is further used to end the timing when the oil fume concentration is greater than the first set concentration during the timing period, and restart the timing until the oil fume concentration is less than the second set concentration again;

[0042] The first monitoring module includes an infrared thermal imager, which obtains image information of the target in the pot and obtains a thermal image based on the image information, thereby obtaining the temperature of the target in the pot;

[0043] The second monitoring module includes an oil fume sensor, and the oil fume sensor monitors the oil fume concentration.

[0044] The above-mentioned intelligent range hood device also includes an image processing module, which is used to analyze the image information obtained by the first monitoring module to determine the food information corresponding to the image information; an food conversion module, which is used to convert the food information into information codes corresponding to nutritional elements and save them in a storage module; an food recommendation module, which is used to recommend food to users using the saved historical information codes, and save the recommended food selected by the user in the storage module; a push module, which is used to recommend recipes to the user client based on the recommended food selected by the user; and a communication module, which is used to communicate with the user client.

[0045] In the above-mentioned intelligent range hood device, a display screen for displaying recipes is also included, and the control module is also used to control the working speed according to the recipe displayed on the display screen;

[0046] The image processing module includes a trained image recognition model, and the training data for the image recognition model includes pictures of various types of food and food information thereof;

[0047] The food conversion module includes a trained food conversion model, and the training data used to train the food conversion model includes information encoding of various food materials and their corresponding nutritional elements;

[0048] The food recommendation module recommends food to the user through steps SB31-SB35.

[0049] The advantages of the present invention are:

[0050] 1. It can realize the timely opening and closing of the range hood to avoid indoor environmental pollution or energy waste caused by users forgetting to open or close it;

[0051] 2. Delayed closing during the process of shutting down the range hood can remove indoor fumes as much as possible, improve the indoor environment, and create a healthier indoor environment;

[0052] 3. Use a two-step monitoring method to start the range hood. After the temperature of the food in the pot has reached the set temperature, the oil fume concentration is monitored and reacted. At this time, the oil fume concentration is relatively high, which can achieve more reliable monitoring results and improve the accuracy and stability of the intelligent control of the range hood;

[0053] 4. Recommending ingredients that users may like and have a balanced nutritional balance based on their eating habits can solve the user's confusion about what to eat, cater to the user's preferences, and improve the nutritional balance of the user's entire eating habits, thereby improving the user's quality of life. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 The module structure frame of the intelligent range hood device Figure 1 ;

[0055] Figure 2 The module structure frame of the intelligent range hood device Figure 2 ;

[0056] Figure 3 is a structural schematic diagram of a range hood in an intelligent range hood device;

[0057] Figure 4 It is a flow chart of intelligent speed control in an intelligent range hood device;

[0058] Figure 5 The process of the intelligent control method in the intelligent range hood device Figure 1 ;

[0059] Figure 6 The process of the intelligent control method in the intelligent range hood device Figure 2 ;

[0060] Figure 7 It is a flow chart of intelligent speed control in an intelligent range hood device;

[0061] Figure 8 A flow chart of a method for recommending food in a smart range hood device.

[0062] Figure numerals: first monitoring module 1; first confirmation module 2; second monitoring module 3; second confirmation module 4; control module 5; image processing module 6; food recommendation module 7; push module 8; display screen 9. DETAILED DESCRIPTION

[0063] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0064] The first aspect of the present invention provides an intelligent range hood device, such as Figure 1 As shown, the device includes a first monitoring module 1, a first confirmation module 2, a second monitoring module 3, a second confirmation module 4 and a control module 5, wherein:

[0065] The first monitoring module 1 uses an image acquisition device to monitor the image information of the ingredients in the pot, including the main ingredients and auxiliary ingredients. The image of the ingredients in the pot is detected, and the temperature in the pot is monitored based on the image information using infrared thermal imaging technology, and the temperature information is transmitted to the first confirmation module 2; the image acquisition device can be installed under the range hood and above the pot surface to acquire the image of the ingredients. In this embodiment, the infrared thermal imager can be used to acquire the image of the ingredients in the pot and directly use infrared thermal imaging technology to monitor the temperature in the pot based on the image information.

[0066] The first confirmation module 2 is used to compare the received temperature information with the set temperature. If the highest temperature monitored in the pot does not reach the set temperature, the range hood is kept in an off state. If the highest temperature monitored in the pot reaches the set temperature, the second monitoring module 3 is notified to enter the monitoring state. The first confirmation module 2 ensures that even if the user is heating food or boiling water, the monitored temperature has reached the preset temperature, but no oil smoke is generated and the range hood will not be turned on. Through two monitorings, the start-up of the range hood will be more accurate, thereby improving the user experience.

[0067] The set temperature can be 60 degrees. When the infrared thermal imager facing the pot through the range hood detects that the temperature is greater than 60℃, the oil fume concentration monitoring is turned on.

[0068] The second monitoring module 3 is used to monitor the concentration of oil smoke around the range hood and absorbed oil smoke. Specifically, the concentration of oil smoke in the air can be monitored by a particle concentration sensor or a VOC gas sensor, the concentration of oil smoke is obtained, and the oil smoke concentration information is transmitted to the second confirmation module 4. The concentration of oil smoke is specifically characterized by the content of oil smoke in the air. For example, the concentration of oil smoke can be the molar ratio of oil smoke to air at the same volume. The concentration of oil smoke can be monitored by directly monitoring the concentration of oil smoke using the particle concentration sensor, or by using the VOC gas sensor to monitor the quality of the air, thereby reflecting the concentration of oil smoke in the air.

[0069] The second confirmation module 4 is used to compare the received oil fume concentration with the first set concentration. If the monitored oil fume concentration is greater than the first set concentration and the range hood is in the off state, the control module 5 is notified to turn on the range hood. The oil fume concentration is continuously monitored when the range hood is in the on state. When the oil fume concentration is less than the second set concentration, the timing starts. After the timing reaches the set time, the control module 5 is notified to turn off the range hood. The first set concentration and the second set concentration can be the same concentration or different concentrations. After the range hood detects that the surrounding oil fume concentration is lower than the second set concentration, the range hood will be turned off after a set time. The set time can be 3 minutes or 5 minutes. Turning off the range hood after the delay setting time can absorb the residual oil fume in the air as much as possible, creating a healthier environment.

[0070] The control module 5 is used to control the opening and closing of the range hood and the operating speed.

[0071] Furthermore, the second monitoring module 3 continues to monitor the oil fume concentration during the timing period, and the second confirmation module 4 is also used to end the timing when the oil fume concentration is greater than the first set concentration during the timing period, and restart the timing until the oil fume concentration is less than the second set concentration again.

[0072] Specifically, my country stipulates that the indoor VOC value shall not be higher than 200g / L, and the VOC requirement must be below 125 grams per liter. Therefore, this solution preferably turns on the range hood when the VOC gas sensor detects that the VOC value is greater than 125g / L; when the VOC gas sensor detects that the VOC value is less than 125g / L, the range hood is automatically turned off after 5 minutes to ensure that the harmful gases in the air are completely absorbed.

[0073] Furthermore, if Figure 2As shown, the device also includes an image processing module 6, a food recommendation module 7, a push module 8, a display screen and a communication module. The image processing module 6 is used to analyze the image information obtained by the first monitoring module 1 to determine the food information corresponding to the image information; the food conversion module is used to convert the food information into information codes corresponding to the nutritional elements and save them in the storage module; the food recommendation module 7 is used to recommend food to the user using the saved historical information codes, and save the recommended food selected by the user in the storage module; the push module 8 is used to recommend recipes to the user based on the recommended food selected by the user; the communication module is connected to the processor of the device, and is used to realize network communication, and to communicate with external databases, servers, user clients, etc., such as obtaining recipes of corresponding ingredients from the database to improve local recipes, obtaining nutritional balance guidelines from the database, receiving recipes selected by the user from the user client and displaying them on the display screen, etc. The food recommendation module 7 pushes the recommended food to the user client or the display screen of the range hood for the user to choose. The user chooses the food. The push module 8 pushes the recommended recipes and the recommended functions of the related food to the user client based on the selected food. The user can choose one or more recipes to save on the user client. When the recipe is needed later, the range hood can choose the recipe. After receiving the user's choice on the user client, the range hood displays the corresponding recipe on the display screen. Then Figure 3 and Figure 4 As shown, the control module automatically adjusts the working speed according to the recipe displayed on the display screen 9. When the recipe recommends a cooking method that produces less oil smoke, such as boiling or steaming, the cooking method is to be operated at a low speed. When the recipe recommends a cooking method that produces more oil smoke, such as stewing or braising, the cooking method is to be operated at a medium speed. When the recipe recommends a cooking method that produces a large amount of oil smoke, such as stir-frying, frying, or grilling, the cooking method is to be operated at a high speed.

[0074] Specifically, the image processing module 6 includes an image recognition model trained by machine learning, and the training data for the image recognition model includes pictures of various types of ingredients and their ingredient information. The trained image recognition model can output ingredient information based on the ingredient pictures.

[0075] The food conversion module includes a food conversion model trained by machine learning, and the training data used to train the food conversion model includes various food information and information codes of corresponding nutritional elements. The trained food conversion model can output information codes of corresponding nutritional elements according to the food information.

[0076] The specific structures and training processes of the image recognition model, food conversion model, and the preference scoring model described below in this solution are not the focus of this solution. Those skilled in the art can select them according to their needs and will not be elaborated here.

[0077] The food recommendation module 7 includes a preference scoring model trained by machine learning. The training data for training the preference scoring model includes historical information encodings of several users and a user-food preference scoring matrix. The trained preference scoring model can output a user-food preference scoring matrix R(m, n) according to the user's historical information encoding, where m represents m users, n represents n kinds of food, and R m，n Represents the preference rating data of user m for ingredient n.

[0078] The food recommendation module 7 analyzes the user's nutritional balance according to the user-food preference score matrix and the nutritional balance guide obtained from the cloud. The closer to the nutritional balance guide requirements, the higher the nutritional balance. The similarity between the target user and other users is calculated according to the user-food preference score matrix, and the adjacent user group of the target user is obtained based on the similarity. The similarity can be calculated using the following similarity algorithms:

[0079] (1) Cosine similarity calculation method: Cosine similarity is used to measure the similarity between individuals by calculating the cosine value of the angle between two vectors in the vector space. The similarity of preferences between users is calculated by solving the cosine value of the angle, that is, the user's food preference is regarded as a point in the n-dimensional coordinate system, and the preference score vector of a certain user is formed by connecting this point with the origin of the coordinate system. The similarity value between two users is the cosine value of the angle between the two preference score vectors. The smaller the angle, the larger the cosine value, which means that the preferences between the two users are more similar. On the contrary, the larger the angle, the smaller the cosine value, which means that the preferences between the two users are more different. In trigonometric coefficients, the range of the cosine value of an angle is between [-1,1]. The cosine value between two overlapping vectors is 1, and the cosine value between two opposite vectors is -1.

[0080] Cosine similarity formula:

[0081]

[0082] Among them, a, b is a point in the n-dimensional coordinate system where the food is located, and the preference score vector of a user is constructed by connecting this point with the origin of the coordinate system.

[0083] For two-dimensional space, assuming ax1, y1, bx2, y2, then:

[0084]

[0085] Among them, x1, y1, x2, y2 are the coordinates of the ingredients in the n-dimensional coordinate system.

[0086] For multidimensional space, suppose aa1,a2,…,a n ,bb1,b2,…,b n ,but:

[0087]

[0088] where a i , b i is the coordinate of the food in n-dimensional space.

[0089] (2) Modified cosine similarity calculation method: Cosine similarity uses cosine distance to find the difference between two individuals. In essence, it focuses on measuring the directional difference of points in space, but is insensitive to numerical differences. Therefore, it is impossible to measure the specific numerical differences in each dimension, which often leads to serious discrepancies between the results and the actual situation. For example, in cluster analysis of e-commerce users, high-value users and low-value users are distinguished by consumption times and average consumption amounts. It is known that the consumption information of users A and B is 4, 20 and 10, 50 respectively. The cosine similarity method can be used to calculate that the similarity between the two users is extremely high, but it is obvious from the numerical value that the value of user B is much higher than that of user A. In order to effectively reduce this error, the modified cosine similarity calculation method was developed, that is, before calculating the cosine similarity between vectors, it is necessary to perform a difference operation between the numerical value on each dimension and the average value of the user's food preference rating. This method has been proven to be more reasonable and in line with reality.

[0090] (3) Pearson correlation coefficient: The Pearson correlation coefficient describes the close relationship between two variables. In the application field of recommendation systems, the correlation coefficient indicates the degree of similarity between two objects. The larger the value, the more similar the objects are.

[0091]

[0092] Among them I i,j represents the food preference rating set reviewed by user i and user j, R i,c represents the preference score of user i for food c, represents the average preference score of user i for the preferred rated food, represents the average preference score of user j for the preferred rated foods.

[0093] (4) Euclidean similarity: a similarity calculation method based on Euclidean distance, which focuses on calculating Euclidean distance, mainly calculating the distance between points in multidimensional space. The concepts of similarity and distance are mutually exclusive to a certain extent. The greater the distance, the smaller the similarity; conversely, the greater the similarity. When solving the similarity of user preference ratings, all food preference ratings that users have jointly evaluated are used as the values ​​of all dimensions in each point, and the straight-line distance between points is calculated. The similarity between two users is reflected by the size of the distance. Assuming that x and y are two points in n-dimensional space, the Euclidean distance between them is:

[0094]

[0095] When n = 2, d(x, y) is the distance between two points on the plane. The similarity calculation based on Euclidean distance is as follows:

[0096]

[0097] Subsequently, after obtaining the neighboring user group of the target user, the food recommendation module 7 selects a number of users whose nutritional balance is higher than the balance threshold from the neighboring user group as the neighbor set of the target user, and then calculates the preference scores of the food ingredients for which the target user has not made a preference score or whose preference scores are lower than the preference score threshold according to the existing preference scores of the neighbor set by the following calculation formula, and sorts them;

[0098]

[0099] Where sim(i,j) represents the similarity between user i and user j, R j,d represents the rating of the nearest neighbor user j on ingredient d, and denote the average ratings of all ingredients given by user i and user j respectively.

[0100] Finally, the N or M ingredients with the highest calculated scores are recommended to the target user. When N < M, N is selected, otherwise M is selected, where M is the number of ingredients for which the ingredient preference scores were calculated previously.

[0101] Specifically, the balance threshold can be a fixed value or determined according to each target user. For example, the nutritional balance of the target user is used as the balance threshold, that is, all other users in the adjacent user group whose nutritional balance is higher than that of the target user are used as the neighbor users of the target user, and food is recommended to the target user based on the neighbor users of the target user. The recommended food is biased towards the preferences of the target user and can improve the nutritional balance of the target user.

[0102] The main function of the food recommendation module 7 is to find neighbor users with similar eating habits and relatively balanced nutrition according to the user's information on the matching of food, and recommend the food with the nutritional elements that the target user lacks in their eating habits to the target user. For example, it is known that the matching of M, N, P, and Q foods by three users ABC is known, among which user A likes to eat two kinds of food A and P, user B prefers N food, and user C likes M, P, and Q. User A's nutritional matching is unbalanced, while user C's nutritional matching is relatively balanced. From the historical records of these users, it can be found that users A and C have similar preferences, so the food Q that user A lacks in user C can be recommended to user A.

[0103] like Figure 3As shown, the first monitoring module 1 is installed outside the range hood, and the second monitoring module 3 can be installed at the entrance of the oil net. When the range hood sucks the indoor oil fume gas into the range hood, the oil fume gas is filtered through the oil net and monitored before the first oil fume separation.

[0104] All the above modules can be connected or embedded in the processor of the range hood. The processor can use MT7688 chip and 128Mbytes DDR memory; it runs embedded Windows operating system and can be installed behind the display screen of the range hood. Figure 3 9 marks are shown.

[0105] The memory of the range hood can store recipe recommendations for various ingredients. The range hood can also be interconnected with an external network through a communication module to obtain recipe recommendations for corresponding ingredients from the external network in real time, or obtain recipe recommendations for corresponding ingredients from the external network to update the recipe recommendations in the memory.

[0106] The communication module can realize communication with the external network through wireless communication via the communication connection of the wireless routing device. The user client can be installed on the user's mobile phone, and the mobile phone communicates with the range hood through wireless signals.

[0107] Specifically, Figure 5 As shown, the second aspect of the present invention provides an intelligent control method for an intelligent range hood:

[0108] SA1. Use infrared thermal imager to monitor the temperature of the target in the pot;

[0109] SA2. When the temperature of the target in the pot exceeds the set temperature, the oil smoke sensor starts to monitor the oil smoke concentration;

[0110] SA3. When the fume concentration is greater than the first set concentration, the range hood is turned on; according to the recipe selected by the user, that is, the recipe displayed on the display screen, or the user manually controls, or according to the image information obtained to determine the current cooking method, and control the range hood to run at low speed, medium speed or high speed according to the cooking method;

[0111] SA4. Continuously monitor the oil fume concentration, and start timing when the oil fume concentration is less than the second set concentration;

[0112] SA5. Turn off the range hood after the set time is reached.

[0113] The above method can timely remove the oil smoke in the air during cooking, and remove the oil smoke remaining in the air after cooking. The user does not need to pay special attention to the range hood, and even if the user forgets to turn on or off the range hood, it can be turned on and off in time, improving the quality of life of the user. In addition, this solution is not that the range hood monitors the oil smoke information in real time and responds, but monitors the oil smoke concentration and responds after the range hood detects that the highest temperature of the food in the pot has reached the preset temperature. At this time, the oil smoke concentration is relatively stable, and the reliability of the monitoring results is high, which can improve the accuracy and stability of the entire range hood intelligent control system.

[0114] Preferably, if Figure 6 As shown, in step SA4, the oil fume concentration continues to be monitored during the timing period. If the oil fume concentration is greater than the first set concentration, the timing is terminated, and the timing is restarted until the oil fume concentration is less than the second set concentration again.

[0115] like Figure 7 As shown, the third aspect of the present invention provides a method for recommending food ingredients for a range hood based on intelligent control of the range hood:

[0116] SB1. During the cooking process with the range hood turned on, an infrared imager is used to obtain image information of the target in the pot, and the image information is analyzed to determine the food information corresponding to the image information;

[0117] SB2. Convert the food information into the corresponding nutritional element information code for storage;

[0118] SB3. Use the saved historical information code to recommend ingredients to users;

[0119] SB4. Save the recommended ingredients selected by the user, and push the recommended recipes to the user client based on the recommended ingredients selected by the user. The user selects one or more recipes to save on the user client, and can select a recipe to be displayed on the range hood display. At this time, when the user is cooking, the range hood controls the working speed according to the recipe preparation method on the display.

[0120] Specifically, Figure 8 As shown, in step SB3, food ingredients are recommended to the user by the following method:

[0121] SB31. Analyze the nutritional balance of users based on the user-ingredient preference scoring matrix; the user-ingredient preference scoring matrix is ​​established based on the user's historical information encoding;

[0122] Calculate the similarity between the target user and other users based on the user-ingredient preference score matrix, and obtain the target user's adjacent user group based on the similarity;

[0123] SB33. Select several users whose nutritional balance degree is higher than the balance threshold from the adjacent user group as the neighbor set of the target user;

[0124] SB34. Calculate the preference scores of ingredients for which the target user has not given a preference score or whose preference scores are lower than the preference score threshold, and sort them according to the existing preference scores of the neighbor set;

[0125] SB35. Recommend the N or M ingredients with the highest calculated scores to the target user. When N<M, take N, otherwise take M, where M is the number of ingredients for which ingredient preference scoring was performed in step SB34.

[0126] Recommending ingredients that users may like and have a balanced nutritional balance based on their eating habits can not only solve the user's confusion about what to eat, but also cater to the user's preferences and improve the nutritional balance of the user's entire eating habits, thereby improving the user's quality of life.

[0127] The specific embodiments described herein are merely examples of the spirit of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in similar ways, but they will not deviate from the spirit of the present invention or exceed the scope defined by the appended claims.

Claims

1. A method for recommending food materials for a range hood based on intelligent control of the range hood, characterized in that: The following methods are included: SB1. During the cooking process with the range hood turned on, obtain the image information of the target in the pot, and analyze the image information to determine the food information corresponding to the image information; SB2. Convert the food information into the corresponding nutritional element information code for storage; SB3. Use the saved historical information code to recommend ingredients to users; SB4. Save the recommended ingredients selected by the user and recommend recipes to the user based on the recommended ingredients selected by the user; In step SB3, food ingredients are recommended to the user by: SB31. Analyze the nutritional balance of users based on the user-ingredient preference scoring matrix; the user-ingredient preference scoring matrix is ​​established based on the preference scoring model trained by machine learning using the user's historical information encoding; Calculate the similarity between the target user and other users based on the user-ingredient preference score matrix, and obtain the target user's adjacent user group based on the similarity; SB33. Select several users whose nutritional balance degree is higher than the balance threshold from the adjacent user group as the neighbor set of the target user; SB34. Calculate the preference scores of ingredients for which the target user has not given a preference score or whose preference scores are lower than the preference score threshold, and sort them according to the existing preference scores of the neighbor set; in Indicates user and users The similarity between Represents the nearest neighbor user About food Ratings, and Respectively represent users and users Average rating for all ingredients; SB35. Recommend the N or M ingredients with the highest calculated scores to the target user. When N<M, take N, otherwise take M, where M is the number of ingredients for which ingredient preference scoring was performed in step SB34.

2. The method for recommending food materials for a range hood based on intelligent control of a range hood according to claim 1, characterized in that: In step SB1, an infrared thermal imager is used to obtain image information of the target in the pot; In step SB4, the recommended recipes are pushed to the user client, and the recommended recipes selected by the user are displayed on the display screen of the range hood.

3. A range hood intelligent control method, characterized in that: The following steps are involved: SA1. Monitor the temperature of the target in the pot; SA2. When the temperature of the target in the pot exceeds the set temperature, start monitoring the fume concentration; SA3. When the oil fume concentration is greater than the first set concentration, the range hood is turned on, and the current cooking method is determined according to the recipe determined by the user based on the recommendation method of claim 1, and the range hood is controlled to operate at a low speed, a medium speed or a high speed according to the cooking method; SA4. Continuously monitor the oil fume concentration, and start timing when the oil fume concentration is less than the second set concentration; SA5. Turn off the range hood after the set time is reached.

4. The intelligent control method for range hood according to claim 3, characterized in that: In step SA4, the oil fume concentration is continuously monitored during the timing period. If the oil fume concentration is greater than the first set concentration, the timing is terminated and the timing is restarted until the oil fume concentration is less than the second set concentration again. In step SA1, the temperature of the target in the pot is monitored by: SA11. Obtain image information of the target in the pot; SA12. Use infrared thermal imaging technology to obtain a thermal image based on the image information, thereby obtaining the temperature of the target in the pot; In step SA2, the oil fume concentration is monitored by an oil fume sensor, and the oil fume sensor is a particle concentration sensor or a VOC gas sensor.

5. An intelligent range hood device, characterized in that: The invention comprises a first monitoring module (1), a first confirmation module (2), a second monitoring module (3), a second confirmation module (4), a control module (5), an image processing module (6), an ingredient conversion module, an ingredient recommendation module (7), a push module (8), and a display screen (9), wherein: A first monitoring module (1) is used to monitor the temperature of a target in the pot and transmit the temperature information to a first confirmation module (2); A first confirmation module (2) is used to compare the received temperature information with the set temperature, and if the monitored temperature information is greater than the set temperature, notify the second monitoring module (3) to enter a monitoring state; A second monitoring module (3) is used to monitor the oil fume concentration and transmit the oil fume concentration information to a second confirmation module (4); A second confirmation module (4) is used to compare the received oil fume concentration with a first set concentration, and if the monitored oil fume concentration is greater than the first set concentration, notify the control module (5) to turn on the range hood; It is also used to continuously monitor the oil fume concentration when the range hood is turned on, start timing when the oil fume concentration is less than a second set concentration, and notify the control module (5) to turn off the range hood after the timing reaches a set time; A control module (5), used to control the on / off and operating speed of the range hood; An image processing module (6) is used to analyze the image information acquired by the first monitoring module (1) to determine the food information corresponding to the image information; The food conversion module is used to convert the food information into the information code of the corresponding nutritional elements and store it in the storage module; An ingredient recommendation module (7) is used to recommend ingredients to the user using the stored historical information code, and to store the recommended ingredients selected by the user in the storage module; the ingredient recommendation module (7) recommends ingredients to the user through steps SB31-SB35 described in claim 1; A push module (8) for recommending recipes to a user client based on the recommended ingredients selected by the user; a communication module for communicating with the user client; The control module (5) is also used to control the operating speed according to the recipe displayed on the display screen.

6. The intelligent range hood device according to claim 5, characterized in that: The temperature information transmitted by the first monitoring module (1) to the first confirmation module (2) is the highest temperature of the target in the pot during the current detection; Alternatively, the temperature information transmitted by the first monitoring module (1) to the first confirmation module (2) is the average temperature of the target in the pot during the current detection; Alternatively, the temperature information transmitted by the first monitoring module (1) to the first confirmation module (2) is the temperature distribution in the current detection, and the first confirmation module (2) selects the highest temperature from the temperature distribution or calculates the average temperature to compare with the set temperature.

7. The intelligent range hood device according to claim 5, characterized in that: The second monitoring module (3) continues to monitor the oil fume concentration within the timing time period, and the second confirmation module (4) is further used to stop timing when the oil fume concentration is greater than the first set concentration within the timing time period, and restart timing when the oil fume concentration is less than the second set concentration again; The first monitoring module (1) comprises an infrared thermal imager, which obtains image information of a target in the pot and obtains a thermal image based on the image information, thereby obtaining the temperature of the target in the pot; The second monitoring module (3) comprises an oil fume sensor, and the oil fume sensor monitors the oil fume concentration.

8. The intelligent range hood device according to claim 5, characterized in that: The image processing module (6) includes a trained image recognition model, and the training data for the image recognition model includes pictures of various types of food ingredients and their food information; The food conversion module (7) comprises a trained food conversion model, and the training data used to train the food conversion model comprises information encoding of various food materials and their corresponding nutritional elements.

Images