Multi-device collaborative food management method and device

Abstract

This application provides a multi-device collaborative food ingredient management method and apparatus, relating to the field of smart home technology. The method includes: acquiring initial food ingredient data; performing data fusion processing on the initial food ingredient data acquired through data acquisition methods corresponding to at least one smart device to obtain target food ingredient data; generating a target recipe in a target format based on the target food ingredient data when a recipe recommendation condition is triggered; wherein the target format is a recommended format corresponding to each smart device; and distributing the target recipe to each smart device so that each smart device recommends the target recipe according to the corresponding recommended format. This application, through collaborative processing of data acquisition, fusion processing, intelligent generation, and multi-device distribution, effectively breaks down the "information silos" between home devices, realizing unified management of food ingredient data across multiple devices and a scenario-adaptive recipe recommendation service.

Description

Technical Field

[0001] This application relates to the field of smart home technology, and in particular to a method and apparatus for multi-device collaborative food management. Background Technology

[0002] With the development of IoT and AI technologies, smart devices such as televisions, smart refrigerators, and screenless speakers have become widely available in homes, each possessing independent data collection and human-computer interaction capabilities. However, in the specific scenario of home food management, these devices often operate independently as "information silos," their capabilities not being effectively integrated. For example, smart refrigerators can collect data but their interaction scenarios are limited; mobile apps offer convenient interaction but rely on cumbersome manual data entry; and televisions and speakers struggle to directly access inventory information. This disconnect between devices and fragmented services prevents users from conveniently and consistently managing and utilizing food information in different living scenarios such as the kitchen and living room, constituting a major obstacle to improving the user experience.

[0003] Currently, most advanced food management solutions are centered around a single device. For example, one solution uses a smart refrigerator equipped with a screen and sensors as its core, managing food through image recognition and weight sensing locally within the refrigerator. Another solution centers on a mobile app, relying on users to actively take photos or manually input data. Both of these typical solutions have inherent flaws: the former confines user interaction firmly to the refrigerator in the kitchen, making remote or cross-scene queries impossible; the latter, due to its complete reliance on user self-discipline and operational burden, results in untimely and inaccurate data updates. The common problem with both is that they fail to leverage the already widespread multi-device ecosystem in the home, and fail to decouple and reorganize data collection, data processing, and information services across different scenarios, thus failing to provide a natural and seamless whole-house food management experience.

[0004] Furthermore, even if users input ingredient information piecemeal across multiple devices, existing technologies lack effective mechanisms to integrate and refine this heterogeneous, multi-source data. Data collected from different devices may conflict (e.g., a refrigerator identifies "tomato" while an app inputs "tomato"), and there's a lack of automatic completion based on a unified knowledge base (e.g., expiration dates). More importantly, recipe recommendation services are often passive and isolated: either users need to actively open specific applications to search, or they rely solely on a single data source (e.g., refrigerator inventory) for simple matching, failing to proactively trigger accurate recommendations based on real-time scenarios such as users watching food programs on TV or asking questions via voice commands. Therefore, existing technological solutions have significant shortcomings in breaking down "information silos" between devices and achieving data fusion and proactive delivery of scenario-based intelligent services. This invention aims to solve this problem. Summary of the Invention

[0005] This application provides a method and apparatus for multi-device collaborative food management. Through collaborative processing of data acquisition, fusion processing, intelligent generation and multi-terminal distribution, it effectively breaks down the "information silos" between home devices and realizes unified management of food data across multiple devices and scene-adaptive recipe recommendation services.

[0006] In a first aspect, this application provides a method for managing food ingredients through multi-device collaboration, comprising the following steps: Acquire initial ingredient data; wherein the initial ingredient data is collected by at least one smart device, and each smart device corresponds to at least one data collection method; The initial ingredient data collected through the data acquisition method corresponding to the at least one smart device is subjected to data fusion processing to obtain the target ingredient data; When the recipe recommendation conditions are triggered, a target recipe in a target format is generated based on the target ingredient data; wherein, the target format is a recommended format corresponding to each smart device; The target recipe is distributed to each smart device so that each smart device can recommend the target recipe according to the corresponding recommendation format.

[0007] Preferably, according to the multi-device collaborative food management method provided in this application, the step of obtaining initial food data includes: Acquire the initial food data collected by the card reader of the target device; Acquire second initial ingredient data collected via the camera of the target device; Acquire input information received through any smart device, and collect third initial ingredient data based on the input information; The target device is a smart device for storing food ingredients, and the input information is obtained by the user through voice and / or clicking input from various smart devices.

[0008] Preferably, according to the multi-device collaborative food management method provided in this application, the step of performing data fusion processing on the initial food data collected through the data acquisition method corresponding to the at least one smart device to obtain target food data includes: The initial ingredient data collected through the data acquisition method corresponding to at least one smart device is subjected to conflict resolution processing to obtain initial standard ingredient data; The initial standard ingredient data is processed by attribute status marking to obtain the target ingredient data.

[0009] Preferably, according to the multi-device collaborative food management method provided in this application, the step of performing conflict resolution processing on the initial food data collected through the data acquisition method corresponding to at least one smart device to obtain initial standard food data includes: When there is a conflict between the same initial ingredient data from different smart devices, the conflict resolution process is carried out according to a preset priority strategy to determine the initial standard ingredient data. The preset priority strategies are arranged from high to low as follows: first initial ingredient data, second initial ingredient data, and third initial ingredient data.

[0010] Preferably, according to the multi-device collaborative food management method provided in this application, the step of performing attribute status marking processing on the initial standard food data to obtain the target food data includes: The initial standard ingredient data is subjected to attribute completion processing to obtain completed ingredient data; The completed ingredient data is processed by state marking to obtain the target ingredient data; The process of completing the ingredient data involves associating attribute information with each initial standard ingredient data according to a preset ingredient attribute database. The attribute information includes at least one of the following: shelf life, storage conditions, recipe information, and cooking difficulty. The target ingredient data involves marking the freshness of each completed ingredient data and initiating a freshness timing process.

[0011] Preferably, according to the multi-device collaborative food management method provided in this application, the triggering conditions for recipe recommendation include at least: timed periodic triggering conditions, device triggering conditions, and food warning triggering conditions; The timed cycle is triggered when the system time reaches the preset cooking time period, which triggers the recipe recommendation process. The device is triggered when any smart device receives a user query command, thus initiating the recipe recommendation process. The food ingredient warning is triggered when the food ingredient reaches its expiration date, which then triggers the recipe recommendation process.

[0012] Preferably, according to the multi-device collaborative food management method provided in this application, the step of generating a target recipe in a target format based on the target food data includes: Based on the target ingredient data and the associated recipe database, determine the corresponding initial recipe; Get the current user's historical recipe flavor data; Based on the historical recipe flavor data and the initial recipe, generate a target recipe in the target format.

[0013] Preferably, according to the multi-device collaborative food management method provided in this application, the step of generating a target recipe in a target format based on the historical recipe flavor data and the initial recipe includes: Based on the historical recipe flavor data, multiple initial recipes are sorted to obtain a flavor recipe sorting list; wherein, the flavor recipe sorting list is a list of multiple initial recipes sorted according to user taste; Based on the freshness priority of the target ingredient data, the target ingredient data is sorted to obtain a list of target ingredient freshness. Generate a corresponding freshness recipe sorting list based on the target ingredient freshness list; Based on the freshness recipe sorting list and the flavor recipe sorting list, a target recipe sorting list is determined, and a target recipe in a target format is determined based on the target recipe sorting list.

[0014] Preferably, according to the multi-device collaborative food management method provided in this application, the step of determining the target recipe in the target format based on the target recipe sorting list includes: Obtain the recommended format for each smart device; Based on the recommended format for each smart device, the target recipe sorting list is converted to obtain the target recipe in the target format.

[0015] Secondly, this application also provides a multi-device collaborative food management device, including the following modules: An acquisition module is used to acquire initial ingredient data; wherein the initial ingredient data is collected by at least one smart device, and each smart device corresponds to at least one data acquisition method. The fusion module is used to perform data fusion processing on the initial food data collected through the data acquisition method corresponding to the at least one smart device to obtain target food data; The generation module is used to generate a target recipe in a target format based on the target ingredient data when the recipe recommendation conditions are triggered; wherein, the target format is a recommended format corresponding to each smart device; The recommendation module is used to distribute the target recipe to each smart device, so that each smart device can recommend the target recipe according to the corresponding recommendation format.

[0016] Thirdly, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the multi-device collaborative food management method as described above.

[0017] Fourthly, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the multi-device collaborative food management method as described above.

[0018] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the multi-device collaborative food management method as described above.

[0019] This application provides a multi-device collaborative food ingredient management method and apparatus. The method involves acquiring initial food ingredient data, which is collected through at least one smart device, with each smart device corresponding to at least one data acquisition method. The initial food ingredient data, collected through the data acquisition methods corresponding to the at least one smart device, undergoes data fusion processing to obtain target food ingredient data. Upon triggering a recipe recommendation condition, a target recipe in a target format is generated based on the target food ingredient data. The target format is a recommended format corresponding to each smart device. The target recipe is then distributed to each smart device, enabling each smart device to recommend the target recipe according to its corresponding recommended format. Through collaborative processing of data acquisition, fusion processing, intelligent generation, and multi-device distribution, the method effectively breaks down the "information silos" between home devices, achieving unified management of food ingredient data across multiple devices and scenario-adaptive recipe recommendation services. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the hardware environment for a multi-device collaborative food management method provided in this application.

[0022] Figure 2 This is one of the flowcharts of the multi-device collaborative food management method provided in this application.

[0023] Figure 3 This is one of the structural schematic diagrams of the multi-device collaborative food management device provided in this application.

[0024] Figure 4 This is the second structural schematic diagram of the multi-device collaborative food management device provided in this application.

[0025] Figure 5This is a schematic diagram of the structure of the electronic device provided in this application. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] The following is combined Figures 1-5 This application describes a multi-device collaborative food management method and apparatus. Through collaborative processing of data acquisition, fusion processing, intelligent generation, and multi-terminal distribution, it effectively breaks down the "information silos" between home devices, and realizes unified management of food data across multiple devices and scene-adaptive recipe recommendation services.

[0028] According to one aspect of the embodiments of this application, a multi-device collaborative food management method is provided. This multi-device collaborative food management method is widely applicable to whole-house intelligent digital control application scenarios such as smart homes, smart home ecosystems, and intelligence house ecosystems. Optionally, in this embodiment, the above-mentioned multi-device collaborative food management method can be applied to, for example... Figure 1 The hardware environment shown consists of terminal device 102 and server 104. For example... Figure 1 As shown, server 104 is connected to terminal device 102 via a network and can be used to provide services (such as application services) to the terminal or clients installed on the terminal. A database can be set up on the server or independently of the server to provide data storage services for server 104. Cloud computing and / or edge computing services can be configured on the server or independently of the server to provide data processing services for server 104.

[0029] The aforementioned network may include, but is not limited to, at least one of the following: wired network, wireless network. The aforementioned wired network may include, but is not limited to, at least one of the following: wide area network, metropolitan area network, local area network. The aforementioned wireless network may include, but is not limited to, at least one of the following: Wi-Fi (Wireless Fidelity), Bluetooth. The terminal device 102 may not be limited to PC, mobile phone, tablet computer, smart air conditioner, smart range hood, smart refrigerator, smart oven, smart stove, smart washing machine, smart water heater, smart washing equipment, smart dishwasher, smart projector, smart TV, smart clothes rack, smart curtains, smart audio-visual equipment, smart socket, smart speaker, smart speaker box, smart fresh air equipment, smart kitchen and bathroom equipment, smart bathroom equipment, smart robot vacuum cleaner, smart window cleaning robot, smart mopping robot, smart air purifier, smart steam oven, smart microwave oven, smart water heater, smart air purifier, smart water dispenser, smart door lock, etc.

[0030] Figure 2 This is one of the flowcharts illustrating a multi-device collaborative food management method provided in this application, such as... Figure 2 As shown, the method may include, but is not limited to, steps S100 to S400: S100, Obtain initial ingredient data; wherein, the initial ingredient data is obtained by collecting data through at least one smart device, and each smart device corresponds to at least one data collection method; S200, perform data fusion processing on the initial ingredient data collected through the data acquisition method corresponding to the at least one smart device to obtain target ingredient data; S300, when the recipe recommendation conditions are triggered, a target recipe in a target format is generated based on the target ingredient data; wherein, the target format is a recommended format corresponding to each smart device; S400, the target recipe is sent to each smart device so that each smart device can recommend the target recipe according to the corresponding recommendation format.

[0031] In step S100 of some embodiments, initial ingredient data is acquired; wherein the initial ingredient data is collected by at least one smart device, and each smart device corresponds to at least one data acquisition method.

[0032] This embodiment details how to use different smart devices in the home to collaboratively complete the initial collection of food data in multiple ways.

[0033] Specifically, in some embodiments of this application, obtaining initial ingredient data includes: Acquire the initial food data collected by the card reader of the target device; Acquire second initial ingredient data collected via the camera of the target device; Acquire input information received through any smart device, and collect third initial ingredient data based on the input information; The target device is a smart device for storing food ingredients, and the input information is obtained by the user through voice and / or clicking input from various smart devices.

[0034] Furthermore, the target device typically refers to a smart refrigerator with storage capabilities. It is equipped with an RFID reader (Radio Frequency Identification reader). When a user places a carton of "fresh milk" with an RFID tag into the refrigerator, the reader automatically scans the tag.

[0035] The RFID tags store standardized information written by the manufacturer. The reader reads this information via radio frequency signals and uploads it to the core service layer through the refrigerator's Wi-Fi module. The data structure read is, for example: {"Name": "Whole Milk", "Production Date": "2023-10-25", "Shelf Life": "7 days", "Brand": "XX"}. Data obtained in this way (i.e., the initial ingredient data) has the advantages of high accuracy and well-structured information.

[0036] This embodiment enables contactless and automated collection of information on pre-packaged ingredients, eliminating the need for any manual input by the user and greatly improving data entry efficiency.

[0037] In some embodiments of this application, for bulk food items without RFID tags, such as a bunch of "spinach" or a few "tomatoes," the built-in camera of the smart refrigerator will capture images of them. The images are transmitted to the refrigerator's local microprocessor, which runs a lightweight AI image recognition model.

[0038] This AI image recognition model is a convolutional neural network (CNN) trained on a large number of food images to identify common fruits, vegetables, meats, etc. After recognition, it generates a second set of initial food data, for example: {"Recognition Result": "Spinach", "Confidence Level": 92%}. The confidence level reflects the model's degree of certainty in its judgment.

[0039] This embodiment solves the problem of automated identification of unpackaged food ingredients and expands the range of manageable food ingredients.

[0040] In some embodiments of this application, users can supplement or correct food information using any convenient smart device.

[0041] For example, by correcting the initial food data through the smart refrigerator's touch screen, after recognizing "spinach", the screen displays the result and allows the user to manually change it to "bok choy" or add "about 300 grams".

[0042] Food information via mobile app: When shopping at the supermarket, users can manually add 2 "steaks" to their "planned purchase" list via the mobile app.

[0043] The initial food data can be corrected using a screenless speaker: Users can input via voice: "Xiao Hai, Xiao Hai, I bought a bag of Wuchang rice and put it in the refrigerator."

[0044] By correcting the initial ingredient data on a smart TV, users can use the remote control to click "Add to My Ingredient Library" on the TV food program interface to add "salmon" from the program to the list.

[0045] Each smart device converts the user's voice (automatic speech recognition to text) or click operation into structured input information, which is then uploaded as the third initial ingredient data.

[0046] This embodiment provides a flexible and diverse data supplementation channel, leveraging the advantages of multi-device interaction to cover the entire process from shopping planning to warehouse confirmation, ensuring data integrity and user engagement.

[0047] In step S200 of some embodiments, the initial ingredient data collected by the data acquisition method corresponding to the at least one smart device is subjected to data fusion processing to obtain target ingredient data.

[0048] This embodiment performs data fusion processing on the initial ingredient data collected through the data acquisition method corresponding to the at least one smart device, illustrating how to process initial data from different sources and with different formats into a unified, accurate, and information-rich "target ingredient data".

[0049] Specifically, the initial ingredient data collected through the data acquisition method corresponding to at least one smart device is subjected to conflict resolution processing to obtain initial standard ingredient data; The initial standard ingredient data is processed by attribute status marking to obtain the target ingredient data.

[0050] More specifically, when there are conflicts in the same initial ingredient data from different smart devices, conflict resolution is performed according to a preset priority strategy to determine the initial standard ingredient data; The preset priority strategies are arranged from high to low as follows: first initial ingredient data, second initial ingredient data, and third initial ingredient data.

[0051] In some embodiments of this application, a user places a bag of food labeled "tomato sauce" with an RFID tag into a refrigerator, while simultaneously manually entering "ketchup" via a mobile app. The system receives both first initial data ("tomato sauce") and third initial data ("ketchup"), which conflict with each other.

[0052] The system executes a preset priority strategy for decision-making. Priorities, from highest to lowest, are: RFID data > AI image recognition data > user-entered data. Based on this rule, the system determines "tomato sauce" as the correct name and generates initial standard ingredient data.

[0053] This strategy is based on the objectivity of the data source and the error probability setting. RFID data comes from the manufacturer and is the most reliable.

[0054] This embodiment automatically resolves the problem of inconsistency between multiple data sources, ensuring the uniqueness and accuracy of core data (such as names) and avoiding data chaos.

[0055] In some embodiments of this application, first initial ingredient data, second initial ingredient data, and third initial ingredient data of the same ingredient can also be compared. When any two initial ingredient data are the same, the abnormal initial ingredient data is corrected based on the two identical initial ingredient data, and the corrected initial ingredient data is used as the initial standard ingredient data.

[0056] More specifically, attribute completion processing is performed on the initial standard ingredient data to obtain completed ingredient data; The completed ingredient data is processed by state marking to obtain the target ingredient data; The process of completing the ingredient data involves associating attribute information with each initial standard ingredient data according to a preset ingredient attribute database. The attribute information includes at least one of the following: shelf life, storage conditions, recipe information, and cooking difficulty. The target ingredient data involves marking the freshness of each completed ingredient data and initiating a freshness timing process.

[0057] It is understood that in some embodiments of this application, after conflict resolution, the ingredient is determined to be "spinach" (from AI recognition). However, the recognition result only includes the name. The system will query the ingredient attribute database maintained by the core service layer.

[0058] This food attribute database is a knowledge base that stores common attributes of frequently used ingredients. The system queries using "spinach" as the key, automatically linking and completing the attribute information to obtain the complete food data, for example: {"Name": "spinach", "Default Shelf Life": "Refrigerated 3 days", "Storage Conditions": "0-4℃", "Category": "Leafy Green Vegetable"}. For shelf life already included in the RFID data, the RFID data will be used and will not be overwritten.

[0059] This embodiment greatly enriches the content of ingredient data, providing key data support for advanced functions such as shelf-life management and recipe matching, and reducing the number of fields that users need to fill in.

[0060] In some embodiments of this application, further, for "spinach," the system calculates the expiration date as October 30th based on its storage date (assumed to be October 27th) and the completed "default shelf life: 3 days refrigerated." The system then marks its status as "fresh" and starts a countdown.

[0061] The system creates a status tracking thread that continuously calculates the difference between the current time and the expiration time. When the difference reaches a warning threshold (e.g., 2 days remaining), the status is automatically updated to "near expiration". If the user clicks "finished" through the app, the status is updated to "consumed".

[0062] This embodiment realizes dynamic and automated management of the food life cycle, transforming static data into dynamic information that can be tracked and alerted, which is the basis for intelligent reminders and accurate recommendations.

[0063] In step S300 of some embodiments, when the recipe recommendation condition is triggered, a target recipe in a target format is generated based on the target ingredient data; wherein, the target format is a recommended format corresponding to each smart device.

[0064] It is understandable that the conditions for triggering recipe recommendations include at least: timed periodic trigger conditions, device trigger conditions, and ingredient warning trigger conditions; The timed cycle is triggered when the system time reaches the preset cooking time period, which triggers the recipe recommendation process. The device is triggered when any smart device receives a user query command, thus initiating the recipe recommendation process. The food ingredient warning is triggered when the food ingredient reaches its expiration date, which then triggers the recipe recommendation process.

[0065] In some embodiments of this application, the system uses a timed, periodic triggering mechanism: A built-in user habit model analyzes and discovers that users typically begin preparing dinner around 6:30 PM on weekday evenings. Therefore, the system automatically triggers a recipe recommendation process every day at 6:25 PM.

[0066] The device is triggered when a user asks the screenless speaker in the living room, "Xiao Hai, how do you cook the potatoes in the fridge?" The speaker uploads this voice query and immediately triggers the recommendation process.

[0067] The food ingredient alert is triggered when, for example, the "spinach" status changes to "near expiration" (1 day remaining), thus triggering a shelf-life warning. Simultaneously, the system also uses this event as one of the "recipe recommendation criteria," automatically initiating a recommendation process to find recipes that utilize "spinach."

[0068] In some embodiments, the system also includes scene-based triggering, such as when a user finishes watching an episode of a food program on a smart TV in the evening, and the program involves "braised pork." The TV reports the end of the viewing. Five minutes later, the user searches for "what meat dishes are recommended" on the refrigerator screen in the kitchen. The system recognizes the behavioral sequence of "searching shortly after watching a food program" and treats it as a high-priority composite trigger condition, giving priority to "braised pork" and related recipes when making recommendations.

[0069] Furthermore, the step of generating a target recipe in a target format based on the target ingredient data includes: Based on the target ingredient data and the associated recipe database, determine the corresponding initial recipe; Get the current user's historical recipe flavor data; Based on the historical recipe flavor data and the initial recipe, generate a target recipe in the target format.

[0070] Understandably, the system takes the current "target ingredient data" (e.g., {"spinach": near expiration, "eggs": plentiful, "rice": plentiful}) as input, searches the recipe database, and determines the corresponding initial recipe.

[0071] Then, a rule-based content filtering algorithm is used to filter out recipes whose ingredients are a subset of the user's inventory. For example, it matches "scrambled eggs with spinach", "scrambled eggs with spinach", and "fried rice with eggs".

[0072] Calculate the ingredient coverage of each recipe (e.g., "scrambled eggs with spinach" covers the near-expiry ingredient "spinach", so the coverage is high).

[0073] In other embodiments of this application, the system abstracts the user's current target food data into a food feature vector. This vector contains not only the food name but also its status and quantity. For example, the target food data ["spinach": {status: "near expiration", quantity: "1 bunch"}, "eggs": {status: "sufficient", quantity: "5"}, "rice": {status: "sufficient", quantity: "1 bowl"}} is preprocessed into a structured set that can be used for rapid comparison.

[0074] Each recipe record in the recipe database contains a list of required ingredients. This list is the set of ingredients that are indispensable to making that recipe. For example, the required ingredients list for the recipe "Spinach and Egg Stir-fry" is ["Spinach", "Eggs"].

[0075] Furthermore, the system filters out recipes whose required ingredient list is a subset of the user's currently available ingredient set. Here, "available ingredient set" typically refers to the set of ingredients in the target ingredient data whose status is not "exhausted" or "expired".

[0076] For each recipe R, calculate the difference between its required ingredient list L_recipe and the user's currently available ingredient set S_user: L_recipe-S_user.

[0077] If the difference set is empty, i.e., L_recipe-S_user=M, it means that the user has all the ingredients required for recipe R, and the recipe is considered a successful match and included in the initial recipe candidate set.

[0078] Example: Based on the target ingredient data above, the available ingredient set S_user = {"spinach", "egg", "rice"}.

[0079] The required list for the recipe "Spinach and Egg Stir-fry" is L1 = {"Spinach", "Eggs"}. L1 - S_user = M, a successful match.

[0080] The required list for the recipe "Fried Rice with Egg" is L2 = {"Egg", "Rice"}. L2 - S_user = M, a successful match.

[0081] The required list for the recipe "Tomato and Egg Soup" is L3 = {"tomato", "egg"}. L3 - S_user = {"tomato"} ≠ M, so the match fails.

[0082] After traversing and matching all recipes, the system obtains an initial set of candidate recipes, such as {"scrambled eggs with spinach", "spinach and egg soup", "fried rice with egg"}. These recipes all meet the basic condition that the user can "make them immediately".

[0083] This embodiment achieves rapid convergence from the global recipe space to the user's feasible recipe space through efficient set operations, ensuring the feasibility of the recommendation results and avoiding recommending recipes that the user cannot make. This is the primary technical guarantee for achieving accurate recommendations.

[0084] Furthermore, after obtaining the initial set of candidate recipes, the system needs to further evaluate the compatibility of each candidate recipe with the user's current ingredient situation (especially near-expiry ingredients). To this end, this invention introduces the quantitative indicator of "ingredient coverage" and designs a refined calculation formula for it.

[0085] The formula for calculating ingredient coverage is defined as follows: Coverage score (Recipe) = Σ(Ingredient weight coefficient_Wi × State influence factor_Si) / N.

[0086] Recipe: Represents a candidate recipe.

[0087] Σ is the summation symbol, indicating that the summation is performed on all the ingredients required for the recipe.

[0088] Ingredient Weight Coefficient (Wi): The weight assigned to the i-th ingredient in the recipe. The core rule is: ingredients nearing their expiration date / urgently needed have a higher weight than ordinary ingredients. For example, we can set: W_nearing date = 1.5, W_sufficient = 1.0, W_missing = 0 (through the aforementioned subset matching, there should be no missing ingredients in the candidate recipes).

[0089] State Influence Factor (Si): This is a factor related to the importance of an ingredient in a specific recipe. For example, ingredients can be divided into "main ingredients" (Si=1) and "auxiliary ingredients / seasonings" (Si=0.5) to reflect that consuming the main ingredient is more valuable than consuming a small amount of seasoning.

[0090] N: Normalization factor, which can be the total number of required ingredients for the recipe, or a fixed constant used to normalize the score to a standard range (such as between 0 and 1) to facilitate comparison between different recipes.

[0091] Assume that in the user's target food data, "spinach" is nearing its expiration date, while "eggs" and "rice" are in sufficient quantity. Weighting is set as follows: W_nearing expiration = 1.5, W_sufficient quantity = 1. For simplicity, let Si = 1 for all foods, and N be the number of food types.

[0092] For the recipe "Spinach and Egg Stir-fry": Required ingredients: Spinach (near expiration), eggs (plenty). Coverage score = (W_spinach) S_spinach + W_egg S_egg) / 2 = (1.5 1.0 + 1.0 1.0) / 2 = 1.25.

[0093] For the recipe "Fried Rice with Egg": Required ingredients: Eggs (sufficient), Rice (sufficient). Coverage score = (1.0) 1.0+ 1.0 1.0) / 2 = 1.

[0094] For the recipe "Spinach and Egg Soup": Required ingredients: spinach (near expiration date), eggs (plenty). Coverage score = (1.5) 1.0 + 1.0 1.0) / 2 = 1.25.

[0095] The calculations above show that the coverage scores of "scrambled eggs with spinach" and "spinach and egg soup" (1.25) are both higher than those of "fried rice with egg" (1). This accurately quantifies the advantage of the first two recipes in prioritizing the consumption of "spinach," a near-expiry ingredient.

[0096] The ingredient coverage score, as a key quantitative indicator, successfully integrates the business objective of "consuming near-expiry ingredients" into the technical algorithm. This provides an objective and calculable data basis for subsequent ranking of recipes based on the freshness list of target ingredients.

[0097] This computational model has good scalability. In some embodiments, more influencing factors can be introduced, such as ingredient costs, nutritional value, and the degree of user aversion to a certain ingredient. These can be easily converted into corresponding weights or factors and incorporated into the formula, thereby enriching the decision-making dimensions of the recommendation system and making it more intelligent.

[0098] The calculated ingredient coverage score will serve as the core ranking criterion for the "Freshness Recipe Ranking List". The system will then combine this score with the "Taste Recipe Ranking List" and use strategies such as weighted fusion to generate the final "Target Recipe Ranking List", thus completing the intelligent decision-making loop from "dishes that can be made" to "dishes most suitable for the current scenario".

[0099] In some embodiments of this application, generating a target recipe in a target format based on the historical recipe flavor data and the initial recipe includes: Based on the historical recipe flavor data, multiple initial recipes are sorted to obtain a flavor recipe sorting list; wherein, the flavor recipe sorting list is a list of multiple initial recipes sorted according to user taste; Based on the freshness priority of the target ingredient data, the target ingredient data is sorted to obtain a list of target ingredient freshness. Generate a corresponding freshness recipe sorting list based on the target ingredient freshness list; Based on the freshness recipe sorting list and the flavor recipe sorting list, a target recipe sorting list is determined, and a target recipe in a target format is determined based on the target recipe sorting list.

[0100] Understandably, the system retrieves the current user's historical recipe taste data (e.g., the user rated "light" recipes highly in the past week and frequently saved "stir-fry" recipes). Simultaneously, it identifies other users with similar tastes within the user group. Analysis reveals that similar users prefer "garlic spinach" to "spinach and egg soup." Therefore, based on the initial recipe list, adjustments are made according to group preferences to obtain a flavor-sorted recipe list that matches the users' tastes.

[0101] Furthermore, the system generates a list of target ingredient freshness, prioritizing "spinach" that is "near its expiration date." Based on this, recipes containing "spinach" receive extra points, generating a freshness-ranked recipe list.

[0102] Furthermore, the system integrates the flavor recipe ranking list and the freshness recipe ranking list, and uses a weighted scoring algorithm to perform a weighted fusion of the freshness recipe ranking list and the flavor recipe ranking list to obtain the target recipe ranking list.

[0103] For example, the weighting of "near-expiry" was set at 70%, and the weighting of the user's "taste" was set at 30%. Ultimately, "scrambled eggs with spinach" (which efficiently consumes near-expiry products and meets the user's preference for quick stir-frying) had the highest overall score and was determined as the first choice in the target recipe ranking list, i.e., the "target recipe".

[0104] In other embodiments of this application, the system maintains a continuously updated user taste preference vector. This vector is obtained by analyzing "historical recipe taste data," which includes, but is not limited to, explicit user ratings (1-5 stars), favorites, clicks to "not recommend again," and feedback after cooking (such as "too salty," "liked").

[0105] The model is constructed using an algorithm that combines collaborative filtering and content analysis. On one hand, the system seeks other user groups with similar rating behaviors to the current user to mine group preferences (user-based collaborative filtering). On the other hand, it analyzes the metadata tags of recipes in the user's personal historical behavior (such as "cuisine: Sichuan", "flavor: spicy", "cooking method: stir-fry", "difficulty: easy") to form a content-based user profile.

[0106] Ultimately, users' taste preferences are quantified into a multi-dimensional vector, for example: User Preference Vector P = {"Spicy": 0.8, "Light": 0.3, "Stir-fry": 0.9, "Soup": 0.5, "Easy Difficulty": 0.95}. The higher the value, the stronger the preference.

[0107] Calculate the match between recipes and user tastes: In the recipe database, each recipe R also has a similar attribute label vector T_R.

[0108] For each recipe in the initial candidate set, calculate the cosine similarity or dot product between its label vector T_R and the user preference vector P, which is used as the taste matching score TasteScore(R) for that recipe.

[0109] Example calculation: Assume that the label for "scrambled eggs with spinach" is {"light": 0.9, "stir-fry": 1, "simple": 1}, and the label for "scrambled eggs with spinach" is {"light": 1.0, "soup": 1, "simple": 1}.

[0110] The flavor matching score for "spinach and scrambled eggs" = similarity score with vector P ≈ 0.3 0.9 + 0.9 1.0+ 0.95 1.0 = 2.12.

[0111] The flavor matching score for "spinach and egg soup" is 0.3. 1.0 + 0.5 1.0 + 0.95 1.0 = 1.75.

[0112] All candidate recipes are sorted in descending order based on their flavor matching scores to generate a sorted list of flavor recipes.

[0113] This step enables the recommendation system to "know what users like," personalizing the general list of feasible recipes and ensuring that the recommended recipes are more likely to match the user's taste habits, thereby improving the user experience and the adoption rate of the recommendation system.

[0114] Furthermore, a freshness-based recipe ranking list is generated to rank the same candidate set based on the urgent consumption needs of ingredients, using another dimension. The specific implementation steps are as follows: First, a freshness list of target ingredients is constructed: the system iterates through the "target ingredient data" and sorts the ingredients according to a preset freshness priority rule. The rule is usually: near-expiry status > sufficient status. For ingredients that are near-expiry, they can be sorted in ascending order of remaining shelf life (the shorter the remaining time, the higher the priority).

[0115] The generated list is an ordered sequence of food ingredient IDs or names, such as: ["Spinach (near expiration - 1 day left)", "Eggs (sufficient)", "Rice (sufficient)". This list clearly defines the food ingredients that need to be consumed first.

[0116] Furthermore, the mapping is to recipe ranking, and this step directly utilizes the ingredient coverage score (R) calculated for each candidate recipe in the above embodiment.

[0117] The ingredient coverage score already contains information about "how well the recipe covers high-priority (near-expiration) ingredients." The higher the score, the more efficient the recipe is at consuming urgent ingredients.

[0118] The system sorts candidate recipes in descending order based on ingredient coverage scores, generating a freshness-ranked recipe list. Using the previous calculation results, the list order is: ["Spinach and Egg Stir-fry": 1.25, "Spinach and Egg Soup": 1.25, "Fried Rice": 1.00] (the first two have the same score and can be further subdivided according to other rules, such as cooking time).

[0119] This step translates the business objectives of "reducing waste and ensuring safety" into a computable ranking logic. It ensures that when making recommendations, the system prioritizes recipes that efficiently address the most pressing food management issues (such as consuming near-expiry products).

[0120] Furthermore, multi-objective fusion is performed to determine the final target recipe ranking list. This step is the core of intelligent recommendation decision-making, requiring a balance between two sometimes conflicting objectives: "user preferences" and "ingredient management." The specific implementation steps are as follows: First, the fusion ranking function is designed: the system adopts a weighted linear combination strategy to calculate a comprehensive recommendation score FinalScore(R) for each candidate recipe.

[0121] The calculation formula is defined as: FinalScore(R) = α Normalized(TasteScore(R)) + β Normalized(CoverageScore(R)) Parameter description: Normalized() is the normalization function. It normalizes the TasteScore and CoverageScore to the [0, 1] interval to eliminate the influence of differences in units and numerical ranges. Common methods include min-max normalization.

[0122] α and β: weighting coefficients, and α + β = 1. These two coefficients are key operational adjustment parameters.

[0123] The β (freshness weight) can be dynamically adjusted based on the triggering scenario. For example, for recommendations triggered by "ingredient alerts," the β value should be significantly increased (e.g., 0.7) to emphasize consuming near-expiry products; for recommendations triggered by users' casual browsing, the α value can be increased (e.g., 0.6) to emphasize taste preferences.

[0124] The weighting can also be slightly adjusted by the user through the app (such as "focusing more on taste when making recommendations" or "focusing more on clearing out inventory").

[0125] For each candidate recipe, obtain its normalized score in the flavor list and freshness list, and substitute it into the above formula to calculate the comprehensive recommendation score FinalScore.

[0126] Example calculation (assuming normalization): Set dynamic weights: α = 0.4, β = 0.6 (this time triggered by the near-term).

[0127] "Spinach and Egg Stir-fry": TasteScore = 0.85, CoverageScore = 1.0, resulting in a FinalScore of 0.4. 0.85 + 0.6 1.0 = 0.94.

[0128] "Spinach and Egg Soup": TasteScore = 0.70, CoverageScore = 1.0, resulting in a FinalScore of 0.4. 0.70 + 0.6 1.0 = 0.88.

[0129] "Egg Fried Rice": TasteScore = 0.90, CoverageScore = 0.6, resulting in a FinalScore of 0.4. 0.90 + 0.6 0.6 = 0.72.

[0130] Sort the recipes in descending order based on the FinalScore, and the resulting list is the target recipe ranking list: ["Spinach and Egg Stir-fry", "Spinach and Egg Soup", "Fried Rice with Egg"].

[0131] This embodiment, by introducing configurable weighting coefficients and normalization processing, achieves a flexible and balanced multi-objective decision-making mechanism. It neither simply caters to users nor mechanically processes ingredients, but rather makes intelligent recommendations that best align with the overall interests of the user (balancing experience and practicality) in different scenarios. This is a significant advancement and innovation compared to single-dimensional recommendation systems. This fused list will serve as direct input for the subsequent "format conversion and distribution" steps.

[0132] In some embodiments of this application, determining the target recipe in the target format based on the target recipe sorting list includes: Obtain the recommended format for each smart device; Based on the recommended format for each smart device, the target recipe sorting list is converted to obtain the target recipe in the target format.

[0133] Understandably, the system first queries the device capability description library maintained by the core service layer to obtain the interactive features declared by each target smart device and the recommended formats it supports. For example, a smart refrigerator reports that it supports the "rich media card" format, which includes images and touch buttons; a mobile terminal APP reports that it supports the "interactive push notification" format, which includes deep links; a smart TV reports that it supports the "video focus card" format; and a screenless speaker reports that it only supports the "voice dialogue text" format. Based on this, the system determines a precise format conversion template for each device.

[0134] Next, based on the aforementioned template, the system performs parallel format conversion processing on the top-ranked target recipes (such as "scrambled eggs with spinach"). For smart refrigerators, the system extracts the recipe thumbnails, key attributes, and operation instructions, encapsulating them into card data packages; for mobile apps, it generates push notifications containing eye-catching prompts and direct links; for smart TVs, it associates high-definition recipe images with preview video streams to generate large-screen card data; for screenless speakers, it transforms the recipe information into a conversational recommendation and designs subsequent dialogue options. Finally, the system outputs a series of data packages with identical content but different formats—the target recipes in the target format—ready to be distributed to the corresponding devices.

[0135] In step S400 of some embodiments, the target recipe is distributed to each smart device so that each smart device recommends the target recipe according to the corresponding recommendation format.

[0136] Understandably, the system, through its core service layer's multi-device collaboration module, distributes the target recipe data packet, which has undergone format conversion, to the optimal path based on the communication capabilities and network status of each smart device. For resident devices like smart refrigerators and screenless speakers, the system uses the low-latency MQTT protocol to publish the data packet to a dedicated topic channel pre-subscribed by the device. For mobile terminal apps, it leverages system-level push services such as FCM or APNs to ensure timely delivery of notifications to users' phones. For smart TVs, it uses stable HTTP long connections or DLNA protocols for high-speed transmission within the home LAN. All distribution processes are accompanied by an acknowledgment mechanism, and the data packet is encrypted and carries a unique serial number to ensure the reliability and security of information transmission.

[0137] Upon receiving the data packet, each smart device immediately invokes its local parsing and rendering engine to execute recommended actions according to the preset recommended format. The smart refrigerator's touchscreen dynamically displays a visually appealing recipe operation card, which users can directly click to view steps or mark as completed. The mobile app displays a high-priority reminder in the system notification bar; clicking it redirects the user to the app's full interactive recipe interface. The smart TV inserts a high-definition video card into its content recommendation bar or standby screensaver, automatically playing a preview video of the recipe. The screenless speaker uses a TTS engine to convert text content into a friendly voice broadcast, then enters listening mode, waiting for the user to select "detailed steps" or "next recommendation" via voice command. Thus, the same recipe recommendation seamlessly integrates into different user scenarios in a form highly adapted to the interactive characteristics of each device.

[0138] In some embodiments of this application, a "local-cloud" dual-caching data synchronization mechanism is also included: When the refrigerator records "spinach" data, it is first stored in the refrigerator's local SQLite database (local cache) and marked as "pending synchronization". At the same time, the refrigerator screen interface is updated immediately, and the user can view it instantly.

[0139] When the home network is restored, the system initiates an incremental synchronization algorithm. Only the data blocks "to be synchronized" are compressed, encrypted, and uploaded to the cloud-based MySQL database. After processing in the cloud, the updates are synchronized to other online devices within the home (such as TVs and mobile apps).

[0140] This embodiment ensures that the data acquisition and basic query functions of the core device (refrigerator) are not affected when the network is unstable or even interrupted, and automatically maintains data consistency of all devices after the network is restored, thereby improving the robustness and response speed of the system.

[0141] In some embodiments of this application, a continuous learning mechanism for user preferences is also included: that is, every time a user clicks on a recommended recipe, saves it, rates it ("too spicy", "very simple"), and marks it as "completed" through the APP after actually cooking it, the system records it.

[0142] For example, the system runs an offline machine learning model periodically (e.g., weekly) and uses the feedback data to retrain the user's preference vector. This vector might be represented as: {"Taste": ["Light": 0.8, "Spicy": 0.2], "Difficulty": ["Easy": 0.9], "Cuisine": ["Chinese": 0.7]}.

[0143] This embodiment transforms the recommendation system from a "one-size-fits-all" approach to a "personalized" one, with the accuracy of recommendations increasing over time, thus achieving truly personalized service.

[0144] The multi-device collaborative food management device provided in this application will be described below. The multi-device collaborative food management device described below can be referred to in correspondence with the multi-device collaborative food management method described above.

[0145] like Figure 3 This is one of the structural schematic diagrams of the multi-device collaborative food management device provided in this application.

[0146] A multi-device collaborative food management system can also include a device terminal layer, a core service layer, and a data storage layer. Specifically, the device terminal layer (data acquisition and interaction entry point) includes: Smart refrigerator (core data acquisition terminal): The structure includes a camera, touch screen, RFID card reader, Wi-Fi module and local microprocessor.

[0147] After the camera captures images of the food, the local microprocessor uses a food identification model to initially identify the type of food. The RFID reader reads the product information (name, shelf life) of the pre-packaged food, and the data is uploaded to the core service layer via the Wi-Fi module.

[0148] The touchscreen allows users to manually edit ingredient information (such as customizing ingredient names and adding expiration dates) and receive recipes and reminders synchronized from the core service layer.

[0149] Mobile App (Core Operation and Data Synchronization Terminal): Includes a food information management module, a recipe collection module, a device binding module, and a message center.

[0150] It enables food image capture and recognition by using the phone's camera, and supports users to manually enter food information; By scanning the QR code of each device through the device binding module, the connection with the core service layer is completed, enabling real-time synchronization of ingredient data and user preferences (collected recipes, dietary restrictions); The message center receives push notifications such as expiration date reminders and recipe recommendations, and allows users to mark ingredients as "consumed" or "expired".

[0151] Smart TV (scenario-based service output terminal): Equipped with Wi-Fi / Ethernet connectivity, voice interaction module, high-definition display screen and HDMI interface.

[0152] The TV accesses the core service layer through a system-level interface to synchronize food data and recipe resources in real time. When a user wants to check the food storage situation at home and think about what dishes can be made, the TV triggers a food matching request. After the core service layer returns a matching recipe, the TV pops up a recipe card in the side bar of the screen. Users can initiate queries via remote control or voice commands (such as "check the vegetables in the refrigerator"). The TV displays ingredient information in text and image format and plays recipe tutorial videos.

[0153] Screenless speaker (voice interaction terminal): Supports Wi-Fi / Bluetooth connection, and has a far-field voice recognition module, audio output module and local buffer unit.

[0154] By accessing the core service layer via MQTT (Message Queuing Telemetry Transport Protocol), users can initiate requests through voice commands (such as "What dishes can I make with potatoes and beef?" or "Remind me that my eggs are about to expire"), and the speaker will convert the voice signal into a text command and send it to the core service layer. It receives text-to-speech (TTS) data returned by the core service layer and provides feedback on ingredients, recipe steps, and expiration date reminders in voice form. It also caches commonly used ingredient data and user preferences locally to ensure that basic queries can be responded to even when the network is offline.

[0155] The core service layer (data processing and collaboration hub) includes: device access module, food information management module, intelligent recommendation module, and multi-device collaboration module. Each module interacts with data through internal interfaces.

[0156] Device access module: Provides a unified communication protocol adaptation interface, supports protocols such as MQTT (for speakers and refrigerators) and HTTP (for apps and TVs), and realizes device identity authentication (based on the device's unique identifier ID and key), encrypted data transmission (using the AES encryption algorithm) and connection status monitoring.

[0157] Before the device is bound for the first time, a unique device identifier and key are generated through the APP; when the device is connected, the identifier and key are submitted to complete the authentication, and a data transmission channel is allocated after the authentication is successful. Monitor device connection status in real time. When a device is offline, cache data to be sent and automatically synchronize it after the connection is restored.

[0158] Ingredient Information Management Module: The core is the ingredient data processing engine, which includes sub-modules for data integration, classification, status updates, and early warning rules.

[0159] The technology is implemented as follows: receiving food data (image recognition results, manually entered information) uploaded by various devices, correcting recognition errors through data fusion algorithms, and determining the final information of the food; storing the data according to the categories of "meat / vegetables / fruits / seasonings" and associating it with the basic attributes of the food (default shelf life, storage conditions). Real-time updates on ingredient status (records the entry time when adding ingredients, and allows users to mark adjustments to the quantity); Preset warning rules (the default is to trigger a reminder 3 days before the expiration date, and users can customize the number of days through the APP) to regularly search for food items that are about to expire.

[0160] The intelligent recommendation module is based on a hybrid recommendation model built on collaborative filtering and content recommendation algorithms. The input parameters include food inventory data, user preferences (APP collection records, historical query logs, and evaluation feedback), cooking difficulty, and seasonal factors.

[0161] The technical implementation involves first using a content recommendation algorithm to filter recipes that match existing ingredients (ingredient matching degree ≥ 80%), and then combining this with a collaborative filtering algorithm to analyze the recipe preferences of similar users and rank the candidate recipes. Recipes containing ingredients that are about to expire are given priority in recommendation, while the cooking difficulty recommendation weight is adjusted based on the user's historical selection (e.g., "easy" difficulty recipes are given priority to novice users). Generate structured data containing recipe name, ingredient quantities, steps, and video links for use by various devices.

[0162] Multi-device collaboration module: Enables service linkage and data synchronization scheduling between devices, including a scene triggering rule engine and a data distribution sub-module.

[0163] The technical implementation pre-sets multiple scene trigger conditions for the scene triggering rule engine (such as "speaker receives 'what to eat today' command → triggers recommendation" and "adds food to refrigerator → synchronizes to all devices"). The data distribution submodule adapts the data format according to the device type (e.g., speakers adapt to voice and text, TVs adapt to text and video links, and apps adapt to text and steps) to ensure the best display effect of the same information on different devices.

[0164] The data storage layer (data persistence and backup) adopts a distributed storage architecture of "local cache + cloud storage". The local cache is deployed in the local storage unit of the smart refrigerator, and the two achieve consistency through a data synchronization engine.

[0165] The technology ensures that real-time data generated by device interactions within the home (such as food search and new records) is stored locally first, so that devices can maintain basic data interaction through the local area network when the network is down. When connected to the internet, local data is uploaded to the cloud via an incremental synchronization algorithm, while simultaneously receiving updated data from the cloud (such as recipe library updates and user operation records on other devices). Ingredient data and user preference data are stored in encrypted form, and cloud data is backed up regularly to prevent data loss.

[0166] like Figure 4 This is the second structural schematic diagram of the multi-device collaborative food management device provided in this application. A multi-device collaborative food management device includes the following modules: The acquisition module 410 is used to acquire initial ingredient data; wherein the initial ingredient data is acquired by at least one smart device, and each smart device corresponds to at least one data acquisition method. The fusion module 420 is used to perform data fusion processing on the initial ingredient data collected through the data acquisition method corresponding to the at least one smart device to obtain target ingredient data; The generation module 430 is used to generate a target recipe in a target format based on the target ingredient data when the recipe recommendation conditions are triggered; wherein, the target format is a recommended format corresponding to each smart device; The recommendation module 440 is used to distribute the target recipe to each smart device, so that each smart device can recommend the target recipe according to the corresponding recommendation format.

[0167] Preferably, the multi-device collaborative food management device provided in this application is specifically used to acquire first initial food data collected by the card reader of the target device; Acquire second initial ingredient data collected via the camera of the target device; Acquire input information received through any smart device, and collect third initial ingredient data based on the input information; The target device is a smart device for storing food ingredients, and the input information is obtained by the user through voice and / or clicking input from various smart devices.

[0168] Preferably, the multi-device collaborative food management device provided in this application is specifically used to perform conflict resolution processing on the initial food data collected through the data acquisition method corresponding to at least one smart device to obtain initial standard food data. The initial standard ingredient data is processed by attribute status marking to obtain the target ingredient data.

[0169] Preferably, the multi-device collaborative food management device provided in this application is specifically used to resolve conflicts based on a preset priority strategy when there are conflicts in the same initial food data from different smart devices, and to determine the initial standard food data. The preset priority strategies are arranged from high to low as follows: first initial ingredient data, second initial ingredient data, and third initial ingredient data.

[0170] Preferably, the multi-device collaborative food management device provided in this application is specifically used to perform attribute completion processing on the initial standard food data to obtain completed food data; The completed ingredient data is processed by state marking to obtain the target ingredient data; The process of completing the ingredient data involves associating attribute information with each initial standard ingredient data according to a preset ingredient attribute database. The attribute information includes at least one of the following: shelf life, storage conditions, recipe information, and cooking difficulty. The target ingredient data involves marking the freshness of each completed ingredient data and initiating a freshness timing process.

[0171] Preferably, the multi-device collaborative food management device provided in this application is specifically used to trigger the recipe recommendation conditions, which include at least: timed periodic trigger conditions, device trigger conditions, and food warning trigger conditions; The timed cycle is triggered when the system time reaches the preset cooking time period, which triggers the recipe recommendation process. The device is triggered when any smart device receives a user query command, thus initiating the recipe recommendation process. The food ingredient warning is triggered when the food ingredient reaches its expiration date, which then triggers the recipe recommendation process.

[0172] Preferably, the multi-device collaborative food management device provided in this application is specifically used to determine the corresponding initial recipe based on the target food data and the associated recipe database; Get the current user's historical recipe flavor data; Based on the historical recipe flavor data and the initial recipe, generate a target recipe in the target format.

[0173] Preferably, the multi-device collaborative food management device provided in this application is specifically used to sort multiple initial recipes according to the historical recipe flavor data to obtain a flavor recipe sorting list; wherein, the flavor recipe sorting list is a list of multiple initial recipes sorted according to user tastes; Based on the freshness priority of the target ingredient data, the target ingredient data is sorted to obtain a list of target ingredient freshness. Generate a corresponding freshness recipe sorting list based on the target ingredient freshness list; Based on the freshness recipe sorting list and the flavor recipe sorting list, a target recipe sorting list is determined, and a target recipe in a target format is determined based on the target recipe sorting list.

[0174] Preferably, the multi-device collaborative food management device provided in this application is specifically used to obtain the recommended format of each smart device; Based on the recommended format for each smart device, the target recipe sorting list is converted to obtain the target recipe in the target format.

[0175] Figure 5 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 5 As shown, the electronic device may include a processor 510, a communications interface 520, a memory 530, and a communication bus 540, wherein the processor 510, communications interface 520, and memory 530 communicate with each other via the communication bus 540. The processor 510 can call logical instructions in the memory 530 to execute a multi-device collaborative ingredient management method. This method includes: acquiring initial ingredient data; wherein the initial ingredient data is collected by at least one smart device, each smart device corresponding to at least one data acquisition method; performing data fusion processing on the initial ingredient data collected by the data acquisition methods corresponding to the at least one smart device to obtain target ingredient data; generating a target recipe in a target format based on the target ingredient data when a recipe recommendation condition is triggered; wherein the target format is a recommendation format corresponding to each smart device; and distributing the target recipe to each smart device so that each smart device recommends the target recipe according to the corresponding recommendation format.

[0176] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0177] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the multi-device collaborative food management method provided by the above methods. The method includes: acquiring initial food data; wherein the initial food data is collected by at least one smart device, and each smart device corresponds to at least one data acquisition method; performing data fusion processing on the initial food data collected by the data acquisition method corresponding to the at least one smart device to obtain target food data; generating a target recipe in a target format based on the target food data when a recipe recommendation condition is triggered; wherein the target format is a recommendation format corresponding to each smart device; and distributing the target recipe to each smart device so that each smart device recommends the target recipe according to the corresponding recommendation format.

[0178] Furthermore, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a multi-device collaborative food management method provided by the above methods. This method includes: acquiring initial food data; wherein the initial food data is collected by at least one smart device, each smart device corresponding to at least one data acquisition method; performing data fusion processing on the initial food data collected by the data acquisition method corresponding to the at least one smart device to obtain target food data; generating a target recipe in a target format based on the target food data when a recipe recommendation condition is triggered; wherein the target format is a recommendation format corresponding to each smart device; and distributing the target recipe to each smart device so that each smart device recommends the target recipe according to the corresponding recommendation format.

[0179] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0180] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0181] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for managing food ingredients through multi-device collaboration, characterized in that, include: Acquire initial ingredient data; wherein the initial ingredient data is collected by at least one smart device, and each smart device corresponds to at least one data collection method; The initial ingredient data collected through the data acquisition method corresponding to the at least one smart device is subjected to data fusion processing to obtain the target ingredient data; When the recipe recommendation conditions are triggered, a target recipe in a target format is generated based on the target ingredient data; wherein, the target format is a recommended format corresponding to each smart device; The target recipe is distributed to each smart device so that each smart device can recommend the target recipe according to the corresponding recommendation format.

2. The multi-device collaborative food management method according to claim 1, characterized in that, The process of obtaining initial ingredient data includes: Acquire the initial food data collected by the card reader of the target device; Acquire second initial ingredient data collected via the camera of the target device; Acquire input information received through any smart device, and collect third initial ingredient data based on the input information; The target device is a smart device for storing food ingredients, and the input information is obtained by the user through voice and / or clicking input from various smart devices.

3. The multi-device collaborative food management method according to claim 2, characterized in that, The step of performing data fusion processing on the initial ingredient data collected through the data acquisition method corresponding to the at least one smart device to obtain target ingredient data includes: The initial ingredient data collected through the data acquisition method corresponding to at least one smart device is subjected to conflict resolution processing to obtain initial standard ingredient data; The initial standard ingredient data is processed by attribute status marking to obtain the target ingredient data.

4. The multi-device collaborative food management method according to claim 3, characterized in that, The process of resolving conflicts in the initial ingredient data collected through at least one smart device to obtain initial standard ingredient data includes: When there is a conflict between the same initial ingredient data from different smart devices, the conflict resolution process is carried out according to a preset priority strategy to determine the initial standard ingredient data. The preset priority strategies are arranged from high to low as follows: first initial ingredient data, second initial ingredient data, and third initial ingredient data.

5. The multi-device collaborative food management method according to claim 3, characterized in that, The step of performing attribute status marking processing on the initial standard ingredient data to obtain the target ingredient data includes: The initial standard ingredient data is subjected to attribute completion processing to obtain completed ingredient data; The completed ingredient data is processed by state marking to obtain the target ingredient data; The process of completing the ingredient data involves associating attribute information with each initial standard ingredient data according to a preset ingredient attribute database. The attribute information includes at least one of the following: shelf life, storage conditions, recipe information, and cooking difficulty. The target ingredient data involves marking the freshness of each completed ingredient data and initiating a freshness timing process.

6. The multi-device collaborative food management method according to any one of claims 1 to 5, characterized in that, The method includes: The trigger conditions for recipe recommendations include at least: timed periodic trigger conditions, device trigger conditions, and food ingredient warning trigger conditions; The timed cycle is triggered when the system time reaches the preset cooking time period, which triggers the recipe recommendation process. The device is triggered when any smart device receives a user query command, thus initiating the recipe recommendation process. The food ingredient warning is triggered when the food ingredient reaches its expiration date, which then triggers the recipe recommendation process.

7. The multi-device collaborative food management method according to any one of claims 1 to 5, characterized in that, The process of generating a target recipe in a target format based on the target ingredient data includes: Based on the target ingredient data and the associated recipe database, determine the corresponding initial recipe; Get the current user's historical recipe flavor data; Based on the historical recipe flavor data and the initial recipe, generate a target recipe in the target format.

8. The multi-device collaborative food management method according to claim 7, characterized in that, The step of generating a target recipe in a target format based on the historical recipe flavor data and the initial recipe includes: Based on the historical recipe flavor data, multiple initial recipes are sorted to obtain a flavor recipe sorting list; wherein, the flavor recipe sorting list is a list of multiple initial recipes sorted according to user taste; Based on the freshness priority of the target ingredient data, the target ingredient data is sorted to obtain a list of target ingredient freshness. Generate a corresponding freshness recipe sorting list based on the target ingredient freshness list; Based on the freshness recipe sorting list and the flavor recipe sorting list, a target recipe sorting list is determined, and a target recipe in a target format is determined based on the target recipe sorting list.

9. The multi-device collaborative food management method according to claim 8, characterized in that, The step of determining the target recipe in the target format based on the target recipe sorting list includes: Obtain the recommended format for each smart device; Based on the recommended format for each smart device, the target recipe sorting list is converted to obtain the target recipe in the target format.

10. A multi-device collaborative food management method device, characterized in that, include: An acquisition module is used to acquire initial ingredient data; wherein the initial ingredient data is collected by at least one smart device, and each smart device corresponds to at least one data acquisition method. The fusion module is used to perform data fusion processing on the initial food data collected through the data acquisition method corresponding to the at least one smart device to obtain target food data; The generation module is used to generate a target recipe in a target format based on the target ingredient data when the recipe recommendation conditions are triggered; wherein, the target format is a recommended format corresponding to each smart device; The recommendation module is used to distribute the target recipe to each smart device, so that each smart device can recommend the target recipe according to the corresponding recommendation format.

Images