Self-service vending machine commodity recommendation method and device and storage medium

Abstract

The invention discloses a self-service vending machine commodity recommendation method and device and a storage medium, and relates to the technical field of self-service vending machines, and the self-service vending machine commodity recommendation method comprises the steps that if it is detected that a cabinet door is opened, commodities taken by a user are determined according to video data collected by a camera; obtaining a keyword corresponding to the taken commodity, and generating a self-adjusting beverage search command based on the keyword; sending the self-adjusting beverage search command to a third-party data platform through a preset interface, receiving a search result fed back by the third-party data platform, and selecting a making course corresponding to a target beverage from the search result; and controlling the display device to display the production course and the commodity recommendation information associated with the production course. The commodity selected by the user is recognized in real time through the camera, the related beverage making course is retrieved according to the keyword of the commodity and pushed to the display device of the self-service vending machine to be displayed, and personalized content recommendation based on the real-time purchase behavior of the user is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of self-service vending machines, and in particular to a method, device, and storage medium for recommending products for self-service vending machines. Background Art

[0002] Traditional self-service vending machines offer only simple product selection and purchase functions, lacking intelligent product recommendations and interactive experiences. Specifically, the product display layout within the vending machine is relatively fixed, and consumers typically search for the products they need based on their own knowledge, without receiving intelligent guidance from the vending machine during the purchase process. This often results in consumers spending a considerable amount of time selecting products, especially for items that require a combination, making it difficult for consumers to quickly find the right combination.

[0003] The above content is only used to assist in understanding the technical solution of this application and does not constitute an admission that the above content is prior art. Summary of the Invention

[0004] The main purpose of this application is to provide a method, device and storage medium for recommending products for self-service vending machines, aiming to solve the technical problem that self-service vending machines cannot provide product recommendations based on users' real-time purchasing behavior.

[0005] To achieve the above objectives, an embodiment of the present application provides a method for recommending products for a self-service vending machine, which is applied to a self-service vending machine. The self-service vending machine is provided with a camera and a display device. The method for recommending products for a self-service vending machine includes: If the cabinet door is detected to be open, the user's pickup item is determined based on the video data collected by the camera; Obtaining keywords corresponding to the pickup product, and generating a self-mixed beverage search command based on the keywords; Sending the self-mixed beverage search command to a third-party data platform through a preset interface, receiving search results fed back by the third-party data platform, and selecting a preparation tutorial corresponding to the target beverage from the search results; The display device is controlled to display the production tutorial and product recommendation information associated with the production tutorial.

[0006] In one embodiment, the steps of sending the self-mixed beverage search command to a third-party data platform via a preset interface, receiving search results fed back by the third-party data platform, and selecting a recipe tutorial corresponding to a target beverage from the search results include: Filtering the search results fed back by the third-party data platform to obtain beverage making tutorials whose keyword matching degree corresponding to the picked-up product is higher than a preset threshold; According to the user's historical purchase records, selecting the beverage making tutorial that matches the beverage category recorded in the historical purchase records as the making tutorial corresponding to the target beverage; If the user has no corresponding historical purchase record, the drink making tutorials are sorted according to the number of clicks on the third-party data platform, and the drink making tutorial with the highest number of clicks is selected as the corresponding making tutorial for the target drink.

[0007] In one embodiment, after sending the self-mixed beverage search command to a third-party data platform via a preset interface, receiving search results fed back by the third-party data platform, and selecting a recipe for a target beverage from the search results, the self-service vending machine product recommendation method further includes: After detecting that the user has completed the purchase operation, obtaining the terminal device identifier of the user; Based on the terminal device identification, the production tutorial is pushed to the user's terminal device.

[0008] In one embodiment, the self-service vending machine further includes a packaging recovery chamber, wherein a packaging identification component is provided in the packaging recovery chamber, and the self-service vending machine product recommendation method further includes: When a package to be recycled is detected in the package recycling chamber, information of the package to be recycled is determined based on the package data collected by the package identification component; If a user account identifier is received, updating the environmental protection points associated with the user account identifier based on the information of the packaging to be recycled; If the user account identifier is not received, the target user account is determined based on historical purchase records, and the environmental protection points corresponding to the target user account are updated.

[0009] In one embodiment, if the user account identifier is not received, the step of determining the target user account based on historical purchase records includes: Obtaining the product identification and recycling time corresponding to the packaging information to be recycled; Query historical purchase records within a preset time period and filter out user accounts that have purchased the product corresponding to the product identifier; If multiple user accounts are screened out, the user account with the smallest time difference between the recycling time and the purchase time of the user account is selected as the target user account; If no matching user account is found, the user account associated with the most recent purchase record is selected as the target user account based on the recycling time.

[0010] In one embodiment, the camera of the self-service vending machine includes a main camera and an auxiliary camera. If the cabinet door is detected to be open, the step of determining the user's pickup of the goods based on the video data collected by the camera includes: Detecting that the cabinet door is opened, obtaining an initial state image of the interior of the self-service vending machine captured by the main camera, and a hand movement trajectory during the cabinet door opening period captured by the auxiliary camera; Detecting that the cabinet door is closed, obtaining a final state image of the interior of the self-service vending machine captured by the main camera; Comparing the final state image with the initial state image to determine the commodity location where the commodity state has changed; Determining the hand's stopping position in front of each of the commodity shelves based on the hand's movement trajectory; The product to be picked up by the user is determined based on the matching relationship between the product location and the stop position.

[0011] In one embodiment, the step of determining the hand's stopping position in front of each of the commodity shelves based on the hand's movement trajectory includes: Dividing the hand movement trajectory into a plurality of sub-trajectory segments according to a preset time window, wherein the sub-trajectory segments correspond to the continuous movement paths of the hand within the commodity storage location; Determining, based on the starting coordinates and the ending coordinates of the sub-trajectory segment, the entry time point of the hand into the commodity storage location and the exit time point of the hand leaving the commodity storage location; Determining a stay time of the hand at the corresponding commodity location according to the entry time point and the exit time point; The dwell time is compared with a preset dwell time threshold to determine the dwell position of the hand in front of each of the commodity shelves.

[0012] In one embodiment, the step of obtaining keywords corresponding to the pickup items and generating a custom drink search command based on the keywords includes: Query the pre-stored product keyword library to obtain the first keyword corresponding to the pickup product; According to the additional features of the pickup product, a second keyword corresponding to the pickup product is obtained; The first keyword and the second keyword are combined according to a preset template to generate the self-mixed beverage search command.

[0013] An embodiment of the present application also provides a self-service vending machine device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the self-service vending machine product recommendation method as described above.

[0014] An embodiment of the present application further provides a storage medium, which is a computer-readable storage medium. A computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the self-service vending machine product recommendation method as described above are implemented.

[0015] One or more technical solutions proposed in this application have at least the following technical effects: This application uses a camera to collect image data after the user opens the cabinet door to determine the goods to be picked up. This changes the limitation of traditional vending machines that only provide product selection and purchase functions, enabling the vending machine to capture the product information actually selected by the user in real time, thereby providing a basis for subsequent product recommendations. Furthermore, this application generates a self-mixed beverage search command based on the keywords of the picked-up goods, and with the help of the rich resources and powerful search capabilities of the third-party data platform, it mines potential demand information related to the user's selected goods, such as suitable drinks, etc., providing users with a richer selection reference and solving the problem that traditional vending machines cannot make personalized associative recommendations based on the goods purchased by the user. In addition, by displaying the corresponding preparation tutorials and related product recommendation information of the selected target drink on the display device of the self-service vending machine, users do not need to filter through a lot of information by themselves, and can intuitively and conveniently obtain the required product combination information and preparation tutorials, effectively saving time and cost, significantly improving the interaction between users and self-service vending machines, and allowing users to obtain more intelligent and efficient guidance and services during the shopping process, thereby greatly improving the overall shopping experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of the first embodiment of the method for recommending products for a self-service vending machine according to the embodiment of the present application; Figure 2 This is a flow chart of a second embodiment of the method for recommending products for a self-service vending machine according to an embodiment of the present application; Figure 3 This is a flow chart of a third embodiment of the method for recommending products for a self-service vending machine according to an embodiment of the present application; Figure 4 This is a flow chart of a fourth embodiment of the method for recommending products for a self-service vending machine according to an embodiment of the present application; Figure 5 This is a structural diagram of the self-service vending machine equipment involved in the embodiment of the present application.

[0017] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0018] It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0019] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0020] Traditional self-service vending machines offer only simple product selection and purchase functions, lacking intelligent product recommendations and interactive experiences. Specifically, the product display layout within the vending machine is relatively fixed, and consumers typically search for the products they need based on their own knowledge, without receiving intelligent guidance from the vending machine during the purchase process. This often results in consumers spending a considerable amount of time selecting products, especially for items that require a combination, making it difficult for consumers to quickly find the right combination.

[0021] In view of the above problems, the present application proposes a method for recommending products for self-service vending machines, which includes: if it is detected that the cabinet door is opened, determining the product that the user picks up based on the video data collected by the camera; obtaining keywords corresponding to the picked-up product, and generating a self-mixed beverage search command based on the keywords; sending the self-mixed beverage search command to a third-party data platform through a preset interface, and receiving the search results fed back by the third-party data platform, and selecting the production tutorial corresponding to the target beverage from the search results; controlling the display device to display the production tutorial, as well as product recommendation information associated with the production tutorial.

[0022] This application uses a camera to collect image data after the user opens the cabinet door to determine the goods to be picked up. This changes the limitation of traditional vending machines that only provide product selection and purchase functions, and enables self-service vending machines to capture the product information actually selected by the user in real time, thereby providing a basis for subsequent product recommendations. Furthermore, this application generates a self-mixed beverage search command based on the keywords of the picked-up goods, and with the help of the rich resources and powerful search capabilities of the third-party data platform, it mines potential demand information related to the user's selected goods, such as suitable drinks, etc., providing users with a richer selection reference and solving the problem that traditional vending machines cannot make personalized associative recommendations based on the goods purchased by the user. In addition, by displaying the corresponding preparation tutorials and associated product recommendation information of the selected target drink on the display device of the self-service vending machine, users do not need to filter through a lot of information by themselves, and can intuitively and conveniently obtain the required product combination information and preparation tutorials, effectively saving time and cost, significantly improving the interactivity between users and self-service vending machines, and allowing users to obtain more intelligent and efficient guidance and services during the shopping process, thereby greatly improving the overall shopping experience.

[0023] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication, and program execution functions, or an electronic device or self-service vending machine capable of implementing the above functions. The following uses a self-service vending machine as an example to illustrate this embodiment and the following embodiments.

[0024] The first embodiment of the self-service vending machine product recommendation method proposed in this application is applied to a self-service vending machine, which is provided with a camera and a display device. Figure 1 The method includes steps S10 to S40: Step S10: If it is detected that the cabinet door is opened, the user's pickup item is determined based on the video data collected by the camera.

[0025] It should be noted that door opening refers to the user opening the door of the self-service vending machine to prepare to pick up the goods. Picking up goods refers to the user actually taking out the goods from the self-service vending machine.

[0026] In this embodiment, with the help of a camera installed inside the self-service vending machine, video and image data are collected in real time, which can accurately capture the user's picking up behavior after the cabinet door is opened, and then determine the products actually taken away by the user.

[0027] In one feasible implementation, a background subtraction method is used to extract the user's hand motion region from video data. This method detects the motion of foreground objects by comparing the differences between consecutive frames. After extracting the hand motion region, an object detection algorithm such as YOLOv5 is used to identify the product being grasped by the hand. Optical flow is then used to track the hand's motion trajectory to determine the hand's dwell time and movements in front of each product location, thereby determining the product actually removed by the user. Finally, product feature matching is used to determine the type of product actually removed.

[0028] It's important to note that tracking hand motion is important because simply identifying the items a user grasps doesn't guarantee the exact items they ultimately remove. A user might pick up multiple items but only take some. By tracking hand motion, we can determine the user's dwell time and movements at different product locations, allowing us to more accurately determine the items they actually remove.

[0029] In another feasible implementation, deep learning algorithms and image recognition are used to determine the items to be picked up. Specifically, when the cabinet door is detected to be open, the user's picking behavior is captured by a camera. First, by analyzing the user's hand movements and trajectory, combined with the changes in the image state after the cabinet door is opened and closed, the actual items picked up by the user when the cabinet door is closed are determined. The image data of the actual items picked up, collected by the camera, is then transmitted to the deep learning model. The deep learning model has been pre-trained with a large number of product images and can automatically identify product features in the image data, match them with a pre-stored product feature library, and output the type of product actually taken by the user.

[0030] As you can understand, this embodiment uses a camera to capture image data after the user opens the door to determine the product to be picked up, breaking through the limitations of traditional vending machines that only provide product selection and purchase functions. By determining the user's picked-up product in real time, it provides a basis for subsequent product recommendations.

[0031] Step S20: Obtain keywords corresponding to the pickup products, and generate a self-mixed beverage search command based on the keywords.

[0032] It should be noted that the keywords corresponding to the picked-up products refer to words that can represent the characteristics of the products actually taken away by the user. These keywords are used for searching and recommending subsequent beverage making tutorials.

[0033] In this embodiment, keywords can be obtained by analyzing information such as product name, product category, and ingredients. The keyword extraction process can also be further optimized by combining information such as the current season and environmental factors. Based on the obtained keywords, a specific template combination can be used when generating a custom drink search command. For example, a custom drink search command can be generated using the template combination of [product name] + [main ingredient] + "drink making tutorial" + [seasonal restrictions] + [product category label].

[0034] For example, if a user buys a bottle of lemon black tea, "lemon" and "black tea" can be directly extracted from the product name as keywords; from the product category point of view, lemon black tea is classified as a tea beverage, so "tea beverage" can also be added as a keyword; in addition, if lemon black tea also contains honey, "honey" can also be used as a keyword; further, if it is currently summer, considering the preference of summer consumers for refreshing and cooling drinks, keywords such as "summer drinks" and "cooling" can also be added. Based on the above keywords, using a preset template combination, a preliminary combination result of "lemon, black tea, honey, drink making tutorial, summer drink, cooling, tea drink" can be obtained. Then, through the text generation and optimization algorithm in natural language processing technology, combined with the language expression habits and search logic in the drink making scenario, the combination results are re-sorted to make the self-mixed drink search command more natural, and finally the self-mixed drink search command of "summer lemon black tea drink making, tea drink making tutorial containing honey" is obtained.

[0035] In one feasible implementation, a keyword mapping table is pre-established for items currently sold in the vending machine. Once the user's item is determined, the corresponding keyword is directly retrieved from the keyword mapping table. This keyword is then combined with a pre-set command template to generate a custom drink search command. Furthermore, for newly appearing items or those not listed in the keyword mapping table, a keyword extraction method based on a word vector model is used to extract descriptive information such as the product name and main ingredients. By analyzing the semantic similarity between the extracted description and known keywords, matching keywords are identified to generate the custom drink search command.

[0036] In another feasible implementation, step S20 includes steps S210 to S230: Step S210: querying a pre-stored product keyword library to obtain a first keyword corresponding to the pickup product.

[0037] It should be noted that the product keyword library pre-stores a large number of common products and their corresponding keywords. Among them, the keywords corresponding to the products are mainly determined in advance by analyzing the basic attributes of the products, such as product name, product category, main ingredients and product functions.

[0038] When building a product keyword library, words that can highlight the characteristics, ingredients, etc. of the product are obtained as keywords and associated with the corresponding products. After the user's pickup product is determined, the corresponding identifier of the pickup product, such as the product barcode, internal number, etc., is used to search and match in the product keyword library, and basic keywords directly related to the product can be quickly obtained. For example, if the pickup product is a bottle of orange juice, according to the product barcode corresponding to the orange juice, the pre-set first keywords such as "juice drink", "orange juice", "vitamin C" corresponding to the orange juice can be obtained.

[0039] Step S220: obtaining a second keyword corresponding to the pickup product according to the additional features of the pickup product.

[0040] Additional features refer to product characteristics beyond basic attributes, primarily including seasonal information and environmental factors. These additional features typically change over time, in different scenarios, and with demand. In this embodiment, based on the additional features of the pickup product, a second keyword corresponding to the pickup product can be derived, enriching the search for "custom beverages."

[0041] Understandably, consumers have different beverage preferences in different seasons and environments. For example, in summer or in hot environments, consumers prefer refreshing drinks to cool down; after exercising, consumers are more likely to seek drinks that can quickly replenish energy and hydration.

[0042] In one feasible implementation, a library of additional keyword features is pre-established based on consumer preferences in different scenarios. For summer or high-temperature environments, secondary keywords such as "summer specialty drinks" and "refreshing" might be associated; for sports and fitness scenarios, secondary keywords such as "energy supplement" and "electrolyte balance" might be associated. By establishing this additional keyword library, a rich and practical secondary keyword library can be provided for custom drink search commands, improving the match between search results and consumer expectations.

[0043] It should be noted that the product keyword library and the additional feature keyword library can be established through manual organization or with the help of machine learning algorithms. Manual organization can be used to organize and summarize keywords corresponding to common products or common seasons and environments. Alternatively, machine learning algorithms can be used to analyze beverage-making tutorials related to various products, or analyze consumer preferences for various products under different additional features. This can reveal the correlations between products and keywords, as well as between products and additional features. Keywords can be supplemented from multiple perspectives, such as basic product attributes and additional features, to enrich the generation dimensions of custom beverage search commands.

[0044] Step S230: combining the first keyword and the second keyword according to a preset template to generate the self-mixed beverage search command.

[0045] The preset template is a pre-designed keyword combination format, and its purpose is to integrate the different keywords obtained in step S210 and step S220 into a sentence structure that conforms to the self-mixed beverage search on the third-party data platform. For example, the preset template is "[first keyword] + [second keyword] + self-mixed beverage making tutorial." The first keyword is "orange juice" and "natural fruit juice", and the second keyword is "summer refreshing drink" and "heat relief". After combining the first keyword and the second keyword according to the preset template, a self-mixed beverage search command of "summer refreshing and heat-relieving self-mixed beverage making tutorial using orange juice and other natural fruit juices as raw materials" can be generated. Through the preset template, scattered keywords are converted into targeted search instructions, the accuracy and effectiveness of the search are improved, and the user is provided with a beverage making tutorial that matches the picked-up goods.

[0046] Step S30: sending the self-mixed beverage search command to a third-party data platform through a preset interface, receiving search results fed back by the third-party data platform, and selecting a preparation tutorial corresponding to the target beverage from the search results.

[0047] It should be noted that the pre-defined interface is a communication interface between the vending machine and the third-party data platform, used for data transmission and exchange. Through the pre-defined interface, the vending machine can send a self-mixed drink search command to the third-party data platform and receive the search results returned by the third-party data platform.

[0048] Additionally, it should be noted that third-party data platforms can be cloud-based servers or databases that provide data such as product information, drink pairings, and recipe tutorials. These platforms possess robust data processing and retrieval capabilities. Based on received search commands for self-mixed drinks, they can match and filter within their own data resources, quickly returning drink recipe tutorials and the required product information related to the items being picked up. Furthermore, third-party data platforms can also be websites with a large library of drink recipe tutorials, or social media platforms with a large user base and rich content sharing. These platforms, with their extensive user data and vibrant community atmosphere, can also provide vending machines with search results that meet user needs.

[0049] It is understandable that since third-party data platforms have richer product information and user data, sending self-mixed beverage search commands to third-party data platforms and receiving search results from the platform can avoid the problem of limited data of the self-service vending machine itself and provide users with more comprehensive, accurate and personalized beverage recommendation services.

[0050] In the first feasible implementation, a multi-condition algorithm is used to process the returned search results. First, unsuitable drink recipes are filtered based on user profile data (such as age and consumption preferences). For example, if the user is young, drink recipes containing alcohol and other ingredients unsuitable for minors are filtered out. Then, the recipes are sorted by popularity, with the most popular drink recipes selected as the target drink recipes. Popularity is calculated using a weighted approach using data such as the number of likes, comments, and page views in the search results.

[0051] In a second possible implementation, real-time environmental data such as the current season and weather is collected and used as auxiliary filtering criteria. For example, in the summer, iced drink recipes may be recommended, while in the winter, hot drink recipes may be recommended. Real-time weather conditions may also be incorporated, such as recommending hot or warming drinks on rainy days, to better tailor recommendations to the user's current needs.

[0052] In a third possible implementation, after receiving search results from a third-party data platform, the product information required for the tutorial in the search results is compared with the vending machine's inventory. Tutorials that include products currently sold by the vending machine are marked, and the search results are reordered. Tutorials that use products currently sold by the vending machine are prioritized for display, improving user convenience. For example, if a user purchases a bottle of lemon tea, and the vending machine's inventory includes honey and ice, tutorials for drinks containing lemon tea, honey, and ice will be prioritized for display.

[0053] In a fourth feasible implementation manner, step S30 further includes steps S31-S32: Step S31: After detecting that the user has completed the purchase operation, obtain the user's terminal device identification.

[0054] It should be noted that the terminal device identifier refers to the identification code of the device used by the user (such as a smartphone), which is used to distinguish different user devices.

[0055] In addition, it should be noted that the purpose of obtaining the terminal device identification is to accurately push the production tutorial to the corresponding user device to improve the user experience.

[0056] Step S32: Based on the terminal device identifier, the production tutorial is pushed to the user's terminal device.

[0057] It should be noted that the pushed content is the production tutorials searched based on the products purchased by the user, which can help the user better utilize the purchased products to further make drinks.

[0058] It is understandable that since users may need preparation guidance after purchasing a product, pushing preparation tutorials for the target drink can help increase user satisfaction.

[0059] Step S40: controlling the display device to display the production tutorial and product recommendation information associated with the production tutorial.

[0060] It should be noted that the display device is a display screen or other display-capable device on a vending machine used to display information. The product recommendation information associated with the recipe tutorial refers to other product information required to make the selected target beverage. The product recommendation information may include products currently on sale in the vending machine.

[0061] In this embodiment, the selected production tutorial is displayed on the display device of the self-service vending machine. At the same time, the product information involved in the production tutorial is obtained, and the pictures and names of these products are organized into product recommendation information and displayed together on the display device.

[0062] Based on the above embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the first embodiment can be referred to the above introduction and will not be described in detail later. Figure 2 , step S30 further includes steps S310 to S330: Step S310: Filter the search results fed back by the third-party data platform to obtain beverage making tutorials whose keyword matching degree corresponding to the pickup product is higher than a preset threshold.

[0063] It should be noted that keyword matching refers to the degree of similarity between the drink-making tutorial in the search results and the keywords corresponding to the product the user is trying to pick up, typically calculated using a text analysis algorithm. The preset threshold is a pre-set value used to determine whether the match meets the criteria.

[0064] In this embodiment, by setting a matching threshold, search results that are irrelevant to user needs can be effectively filtered out, ensuring that the recommended beverage making tutorials are more targeted.

[0065] It is understandable that since user needs are usually related to purchasing goods, making tutorials through keyword matching and screening can avoid recommending content that is irrelevant to the goods to be picked up to users, thereby improving the accuracy of recommendations and user experience.

[0066] Step S320: According to the historical purchase records of the user, the beverage making tutorial that matches the beverage category recorded in the historical purchase records is selected as the making tutorial corresponding to the target beverage.

[0067] It should be noted that a user's purchase history refers to detailed information about the user's past purchases, including the type of goods purchased, the time of purchase, etc. By analyzing the purchase history, the user's preferences can be understood.

[0068] In this embodiment, the beverage making tutorial that matches the historical purchase record is selected to predict the beverage making direction that the user may be interested in based on the user's historical purchase behavior.

[0069] It is understandable that since a user's historical purchasing behavior can reflect user preferences, recommending beverage making tutorials based on historical records can avoid recommending product information that the user is not interested in, thereby improving the relevance of recommendations and user satisfaction.

[0070] Step S330: If the user has no corresponding historical purchase record, the beverage making tutorials are sorted according to the number of clicks on the third-party data platform, and the beverage making tutorial with the highest number of clicks is selected as the corresponding making tutorial for the target beverage.

[0071] It should be noted that the number of clicks refers to the number of times the drink-making tutorial is clicked and viewed on a third-party data platform, reflecting the popularity of the drink-making tutorial.

[0072] In addition, it should be noted that when the user has no historical purchase records to refer to, the number of clicks is an effective alternative indicator that can help the vending machine select the most popular beverage making tutorials for recommendation.

[0073] It is understandable that since drink-making tutorials with high click-through rates are usually more popular, selecting drink-making tutorials in this way can avoid inaccurate recommendations caused by a lack of user personal data, thereby improving the universality of recommendations.

[0074] In one feasible implementation, drink-making tutorials can also be ranked based on the number of likes, comments, or popularity on third-party data platforms. Popularity can be calculated by weighting data such as likes, comments, and pageviews. This approach allows for a more comprehensive assessment of the popularity of drink-making tutorials, thereby providing users with more valuable tutorial recommendations.

[0075] The third embodiment of the self-service vending machine product recommendation method proposed in this application is the same or similar to the first embodiment mentioned above. Please refer to the above introduction and will not be repeated hereafter. On this basis, the self-service vending machine also includes a packaging recycling chamber, in which a packaging identification component is provided. Please refer to Figure 3 The method includes steps S50 to S70: Step S50: When a package to be recycled is detected in the package recycling chamber, information of the package to be recycled is determined based on the package data collected by the package identification component.

[0076] It's important to note that the packaging recycling chamber is the space in the vending machine where users place product packaging. It's designed to encourage packaging recycling and enhance environmental awareness. The packaging recognition component is a sensor or scanning device installed within the packaging recycling chamber that identifies and captures the product category of the recycled packaging. Recyclable packaging information refers to the product category of the recycled packaging, as captured and analyzed by the packaging recognition component. This information is then used to determine the environmental credits associated with the recycled packaging.

[0077] In this embodiment, the self-service vending machine classifies and processes the packaging to be recycled, collects data and determines the information of the packaging to be recycled through the packaging recognition component, thereby avoiding errors in manual recognition and improving the efficiency and accuracy of packaging recycling.

[0078] Step S60: If the user account identifier is received, the environmental protection points associated with the user account identifier are updated based on the information of the packaging to be recycled.

[0079] It should be noted that the user account ID refers to the user's identity at the vending machine, which is used to record the user's purchasing behavior, points, and other information. Environmental points are points earned by participating in packaging recycling, which can be redeemed for products or discounts. Updating environmental points involves calculating and adjusting the balance of environmental points in the user's account based on the points associated with the recycled packaging.

[0080] It is understandable that by updating environmental protection points, users can be encouraged to actively participate in packaging recycling and improve their environmental awareness.

[0081] Step S70: If the user account identifier is not received, the target user account is determined based on historical purchase records, and the environmental protection points corresponding to the target user account are updated.

[0082] It should be noted that historical purchase records refer to the user's previous purchase order data at the vending machine, including information such as purchase time and type of purchased goods. Through historical purchase records, the user account associated with the current packaging information to be recycled can be identified.

[0083] In addition, it should be noted that the target user account refers to the user account that best matches the current packaging information to be recycled, as matched through historical purchase records.

[0084] In this embodiment, some users may not actively provide their account number when recycling packaging. Therefore, determining the target user account number and updating the environmental protection points through historical purchase records can avoid the inability to record environmental protection points due to the user not providing an account number, thereby improving the flexibility of environmental protection point updates and user experience.

[0085] In one feasible implementation, users must first complete account verification before the recycling chamber door automatically opens. Users can then place the recycled packaging into the recycling chamber to update their environmental points. By requiring users to provide their account information before recycling, each recycling operation is accurately linked to the corresponding user account, eliminating uncertainty in the allocation of environmental points.

[0086] In a feasible implementation, step S70 further includes steps S710 to S740: Step S710: Obtain the product identification and recycling time corresponding to the packaging information to be recycled.

[0087] It should be noted that the product identifier refers to the product identification code corresponding to the recycled packaging, such as a product barcode or QR code, which is used to help identify the type of product to be recycled. The recycling time refers to the specific time when the recycled packaging is placed in the packaging recycling chamber.

[0088] Step S720: query historical purchase records within a preset time period, and filter out user accounts that have purchased the product corresponding to the product identifier.

[0089] It should be noted that the preset time period refers to a time range set according to actual matching requirements, such as within 3 hours or within the same day, which is used to limit the scope of querying historical purchase records.

[0090] Without a user account ID, packaging recycling must be linked to the user's past purchases. After obtaining the product ID and recycling time, the user's purchase history can be determined by querying the purchase history. Based on the user account identified as having purchased the product, the environmental points for that user account are updated.

[0091] It is understandable that since users may purchase the same product at different times, filtering by limiting historical purchase records within a preset time period can avoid mismatching due to excessive time, thereby improving the accuracy and efficiency of user account matching.

[0092] Step S730: If multiple user accounts are screened out, the user account with the smallest time difference between the recycling time and the purchase time of the user account is selected as the target user account.

[0093] It should be noted that the time difference refers to the interval between the recovery time and the purchase time, usually calculated in hours or days.

[0094] If multiple user accounts have purchased the same product, comparing the time difference between the recycling date and the purchase date can determine which user account's purchase behavior is most closely related to the current recycling behavior. Selecting the user account with the smallest time difference as the target user account allows for more accurate matching of recycling behaviors, thereby improving the accuracy of environmental points updates.

[0095] It is understandable that since users usually recycle packaging within a short period of time after purchasing goods, selecting the user account with the smallest time difference as the target user account can avoid misjudgment caused by too long time intervals, thereby improving the accuracy of the system in identifying the target user account.

[0096] In one feasible implementation, a minimum time difference threshold is set. When multiple user accounts are screened out that have purchased the same product, candidate user accounts whose time difference between the purchase time and the recycling time is greater than the minimum time difference threshold are first screened out. The user account with the smallest time difference is then selected from the candidate user accounts as the target user account. By setting a minimum time difference threshold, misjudgments caused by too short a time difference can be effectively avoided, and user accounts with time differences within a reasonable range can be preferentially selected as target user accounts, ensuring the fairness of environmental protection points updates. For example, within 1 minute, user A purchased product C, and user B also recycled the packaging of product C. At this time, if the target user account is determined based on the minimum time difference, the environmental protection points may be allocated to the account corresponding to user A.

[0097] Step S740: If no matching user account is found, the user account associated with the most recent purchase record is selected as the target user account based on the recycling time.

[0098] It should be noted that the most recent purchase record refers to the record of the last purchase of goods recorded before the recycling time.

[0099] In this way, it can be ensured that even if there is no clear match, a closest user account can be found to update the environmental protection points.

[0100] It is understandable that since it may not be possible to accurately match a specific user account in some cases, selecting the user account associated with the most recent purchase record can avoid the inability to allocate environmental protection points due to the lack of a clear match.

[0101] Based on the above embodiments of the present application, in the fourth embodiment of the present application, the same or similar contents as those in the first embodiment can be referred to the above introduction, and no further details will be given later. On this basis, the camera of the self-service vending machine also includes a main camera and an auxiliary camera, please refer to Figure 4 , step S10 further includes steps S110 to S150: Step S110: Detecting that the cabinet door is opened, obtaining an initial state image of the interior of the self-service vending machine captured by the main camera, and a hand movement trajectory during the cabinet door opening period captured by the auxiliary camera.

[0102] It should be noted that the main camera is installed inside the vending machine to capture the entire display of merchandise, recording the merchandise layout before and after the door is opened. The auxiliary camera is used to capture the user's hand movements and movement trajectory, recording the user's hand operations during the pickup process. The initial state image refers to the image of the merchandise layout inside the vending machine when the door is open, reflecting the merchandise status before the user begins to pick up the merchandise. The hand movement trajectory refers to the movement path of the user's hand while the door is open. By analyzing hand movements, the product location touched by the user can be determined.

[0103] Step S120: Detecting that the cabinet door is closed, obtaining a final state image of the interior of the self-service vending machine captured by the main camera.

[0104] It should be noted that the final state image refers to the product layout image inside the self-service vending machine when the cabinet door is closed, reflecting the status of the products after the user completes picking up the products.

[0105] Step S130: Compare the final state image with the initial state image to determine the commodity location where the commodity state has changed.

[0106] It should be noted that a change in the state of a product refers to a change in the position, quantity or existence of the product between the initial state image and the final state image, for example, the product is taken away or put back.

[0107] The comparison process can be achieved through an image recognition algorithm, which can automatically detect changes in the goods in the image and determine the specific location of the goods that have changed.

[0108] It is understandable that since the user may touch multiple products during the pickup process, but ultimately only take away some of the products, by comparing the initial state image and the final state image, the location of the products actually taken away by the user can be accurately identified, thereby improving the accuracy of identifying the picked-up products.

[0109] In one feasible implementation, the location of products whose status has changed is determined based on image segmentation and feature matching. Specifically, an image segmentation algorithm is first used to segment the initial and final image states, extracting the outline and location information of each product. A feature matching algorithm is then used to extract the initial feature points of the product in the initial image and the final feature points in the final image. The initial and final feature points are then matched. Based on the matching results, the location change of the product is determined. If the feature points corresponding to a product disappear in the final image, the product is considered to have changed its status.

[0110] In another feasible implementation, the ratio of the number of matching feature points to the total number of feature points can be calculated as a similarity index for the change in the state of the product. Specifically, a feature extraction algorithm is first used to extract the initial feature points of the product in the initial state image and the final feature points of the product in the final state image. Next, the initial feature points are matched with the final feature points, and the number of successfully matched feature points (i.e., the number of matching feature points) and the total number of feature points extracted from the initial state image (i.e., the total number of feature points) are counted. Finally, the similarity index between the initial state image and the final state image is calculated by calculating the ratio of the number of matching feature points to the total number of feature points. If the similarity index is lower than a preset similarity threshold, it indicates that the state of the product has changed significantly, such as the product being taken away; if the similarity index is higher than or equal to the preset similarity threshold, it is considered that the state of the product has not changed significantly.

[0111] Step S140: determining the stopping position of the hand in front of each of the commodity shelves according to the hand movement trajectory.

[0112] It should be noted that the dwell position refers to the specific position where the user's hand dwells in front of a certain commodity shelf while the cabinet door is open. The dwell position and dwell time of the hand can be determined by analyzing the hand movement trajectory.

[0113] It is understandable that since users may touch multiple items during the pickup process but ultimately only take away some of the items, by analyzing the hand movement trajectory and stop position, the user's actual pickup behavior can be more accurately inferred, thereby improving the accuracy of picking up items.

[0114] In a feasible implementation, step S140 further includes steps S1410 to S1440: Step S1410: Divide the hand movement trajectory into a plurality of sub-trajectory segments according to a preset time window, wherein the sub-trajectory segments correspond to the continuous movement paths of the hand within the commodity storage location.

[0115] It should be noted that the preset time window refers to a time interval set according to actual needs, which is used to segment the continuous hand movement trajectory into sub-trajectory segments within multiple time periods. A sub-trajectory segment is the continuous movement path of the hand within the preset time window, reflecting the hand's movement within a specific time period.

[0116] By dividing the hand movement trajectory into multiple sub-trajectory segments, the movement characteristics of the hand in different time periods can be refined, thereby more accurately identifying the hand's behavior in front of each product shelf.

[0117] In one feasible implementation, starting from the start time of the hand movement trajectory data, the data is divided according to a preset time window, and the hand movement path within each time period is a sub-trajectory segment. During the segmentation process, the current time point of the segmentation is continuously recorded. When the time period between the start time and the current time point exceeds the preset time window, the start time is updated to the current time point, and the recording of the next sub-trajectory segment begins. For the last time period, if the remaining time is less than the preset time window, it can be divided into a separate sub-trajectory segment or directly discarded according to actual needs.

[0118] After the division is completed, the sub-trajectory segment is verified. By checking whether the distance of the hand movement between adjacent time points in the sub-trajectory segment is reasonable, it is determined whether the sub-trajectory segment actually corresponds to the continuous movement path of the hand in the commodity shelf. Specifically, if the distance between a certain time point and the previous time point in a sub-trajectory segment exceeds the preset distance threshold, the breakpoint position between the two time points is determined. Then, the relative position of the breakpoint position in the entire sub-trajectory segment is calculated. If the breakpoint position is close to the starting position or the ending position of the sub-trajectory segment, the trajectory segment near the breakpoint position is divided into the adjacent sub-trajectory segment. For example, if the breakpoint position is close to the ending position of the sub-trajectory segment, the part of the trajectory after the breakpoint position is divided into the next sub-trajectory segment; if the breakpoint position is close to the starting position of the sub-trajectory segment, the part of the trajectory before the breakpoint position is divided into the previous sub-trajectory segment.

[0119] The divided sub-trajectory segments are associated with product locations. Based on the hand coordinate information in each sub-trajectory segment and the corresponding location range of the product location, each sub-trajectory segment is determined to be within a specific product location. The degree of match between the coordinates of all time points within the sub-trajectory segment and the coordinate range of the product location is calculated. If the match exceeds a preset threshold, the sub-trajectory segment is considered to correspond to the product location.

[0120] Step S1420: Determine the entry time point of the hand into the commodity storage location and the exit time point of the hand from the commodity storage location based on the starting coordinates and the ending coordinates of the sub-trajectory segment.

[0121] It's important to note that the starting coordinates refer to the hand's position when it enters a product location, and the ending coordinates refer to the hand's position when it leaves a product location. By analyzing the starting and ending coordinates of each sub-trajectory segment, we can calculate the time points when the hand enters and leaves the product location, respectively, and determine the hand's stay in front of each product location.

[0122] Step S1430: Determine the stay time of the hand at the corresponding commodity location according to the entry time point and the exit time point.

[0123] It should be noted that the dwell time refers to the time interval between the time a hand enters and leaves a certain product location. By calculating the time difference between the entry and exit times, the dwell time of the hand in front of the product location can be obtained.

[0124] Step S1440: Compare the dwell time with a preset dwell time threshold to determine the dwell position of the hand in front of each of the commodity shelves.

[0125] It should be noted that the preset dwell time threshold is used to determine whether the hand's stay in front of the product shelf is meaningful. If the dwell time is greater than or equal to the preset threshold, it is considered that the hand has a significant dwelling behavior in front of the product shelf, and a pickup operation may occur.

[0126] In this embodiment, since the user may briefly stop in front of multiple product locations during the pickup process, by comparing the stop time with the preset stop time threshold, the product locations that the user is really interested in can be screened out, avoiding misjudgments caused by short stops, thereby improving the accuracy of picking up product identification.

[0127] Step S150: Determine the product to be picked up by the user based on the matching relationship between the product location and the stop position.

[0128] It should be noted that the matching relationship refers to determining the product location when the user actually picks up the product by comparing the product location with the rest position of the product whose status has changed, and then determining the product that the user picks up based on the product category corresponding to the product location.

[0129] An embodiment of the present application provides a self-service vending machine device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the self-service vending machine product recommendation method in the above-mentioned embodiment one.

[0130] Reference below Figure 5, which shows a schematic diagram of the structure of a self-service vending machine device suitable for implementing the embodiment of the present application. The self-service vending machine device in the embodiment of the present application may include various hardware and software components for implementing the self-service vending machine product recommendation method. Figure 5 The self-service vending machine device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0131] like Figure 5 As shown, the vending machine device may include a processing device 1001 (e.g., a central processing unit, graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of the vending machine device. Processing device 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to I / O interface 1006: input devices 1007, such as a touchscreen, touchpad, or keyboard; output devices 1008, such as a liquid crystal display (LCD), speaker, or vibrator; storage device 1003, such as a magnetic tape or hard disk; and communication devices 1009. The communication device 1009 can allow the vending machine device to communicate with other devices wirelessly or wired to exchange data. Although the figure shows a vending machine device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0132] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a read-only memory 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are performed.

[0133] The self-service vending machine device provided in this application utilizes the self-service vending machine product recommendation method described in the aforementioned embodiment, resolving the technical issue of self-service vending machines being unable to provide product recommendations based on users' real-time purchasing behavior. Compared to the prior art, the self-service vending machine device provided in this application achieves the same beneficial effects as the self-service vending machine product recommendation method described in the aforementioned embodiment. Other technical features of this self-service vending machine device are the same as those disclosed in the aforementioned embodiment and are not further elaborated upon here.

[0134] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0135] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

[0136] An embodiment of the present application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, and the computer-readable program instructions are used to execute the self-service vending machine product recommendation method in the above embodiment.

[0137] The computer-readable storage medium provided herein may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including, but not limited to, wires, optical cables, radio frequency (RF), etc., or any suitable combination thereof.

[0138] The computer-readable storage medium may be included in the self-service vending machine device; or it may exist independently without being assembled into the self-service vending machine device.

[0139] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the self-service vending machine device, the self-service vending machine device: if it detects that the cabinet door is opened, determines the user's pickup goods based on the video data collected by the camera; obtains the keywords corresponding to the pickup goods, and generates a self-mixed beverage search command based on the keywords; sends the self-mixed beverage search command to a third-party data platform through a preset interface, receives the search results fed back by the third-party data platform, and selects the production tutorial corresponding to the target beverage from the search results; controls the display device to display the production tutorial and the product recommendation information associated with the production tutorial.

[0140] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0141] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0142] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0143] The computer-readable storage medium provided in this application stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned method for recommending products for a self-service vending machine. This computer-readable storage medium can address the technical issue of self-service vending machines being unable to provide product recommendations based on users' real-time purchasing behavior. Compared to the prior art, the beneficial effects of the computer-readable storage medium provided in this application are similar to those of the method for recommending products for a self-service vending machine provided in the aforementioned embodiment, and are not further elaborated here.

[0144] An embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned self-service vending machine product recommendation method.

[0145] The computer program product provided in this application can address the technical issue of self-service vending machines being unable to provide product recommendations based on users' real-time purchasing behavior. Compared to the prior art, the beneficial effects of the computer program product provided in this embodiment are the same as those of the self-service vending machine product recommendation method provided in the aforementioned embodiment, and are not further elaborated here.

[0146] The above are only preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent processing scope of the present application.

[0147] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0148] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method.

[0149] The above are only preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for recommending products for a self-service vending machine, characterized in that: Applied to a self-service vending machine, the self-service vending machine is provided with a camera and a display device, and the product recommendation method of the self-service vending machine includes: If the cabinet door is detected to be open, the user's pickup item is determined based on the video data collected by the camera; Obtaining keywords corresponding to the pickup product, and generating a self-mixed beverage search command based on the keywords; Sending the self-mixed beverage search command to a third-party data platform through a preset interface, receiving search results fed back by the third-party data platform, and selecting a preparation tutorial corresponding to the target beverage from the search results; The display device is controlled to display the production tutorial and product recommendation information associated with the production tutorial.

2. The method for recommending products for a self-service vending machine according to claim 1, wherein: The steps of sending the self-mixed beverage search command to a third-party data platform through a preset interface, receiving search results fed back by the third-party data platform, and selecting a preparation tutorial corresponding to a target beverage from the search results include: Filtering the search results fed back by the third-party data platform to obtain beverage making tutorials whose keyword matching degree corresponding to the picked-up product is higher than a preset threshold; According to the user's historical purchase records, selecting the beverage making tutorial that matches the beverage category recorded in the historical purchase records as the making tutorial corresponding to the target beverage; If the user has no corresponding historical purchase record, the drink making tutorials are sorted according to the number of clicks on the third-party data platform, and the drink making tutorial with the highest number of clicks is selected as the corresponding making tutorial for the target drink.

3. The method for recommending products for a self-service vending machine according to claim 1, wherein: After the steps of sending the self-mixed beverage search command to a third-party data platform via a preset interface, receiving search results fed back by the third-party data platform, and selecting a recipe for a target beverage from the search results, the self-service vending machine product recommendation method further includes: After detecting that the user has completed the purchase operation, obtaining the terminal device identifier of the user; Based on the terminal device identification, the production tutorial is pushed to the user's terminal device.

4. The method for recommending products for a self-service vending machine according to claim 1, wherein: The self-service vending machine further includes a packaging recovery chamber, wherein a packaging identification component is provided in the packaging recovery chamber. The self-service vending machine product recommendation method further includes: When a package to be recycled is detected in the package recycling chamber, information of the package to be recycled is determined based on the package data collected by the package identification component; If a user account identifier is received, updating the environmental protection points associated with the user account identifier based on the information of the packaging to be recycled; If the user account identifier is not received, the target user account is determined based on historical purchase records, and the environmental protection points corresponding to the target user account are updated.

5. The method for recommending products for a self-service vending machine according to claim 4, wherein: If the user account identifier is not received, the step of determining the target user account based on historical purchase records includes: Obtaining the product identification and recycling time corresponding to the packaging information to be recycled; Query historical purchase records within a preset time period and filter out user accounts that have purchased the product corresponding to the product identifier; If multiple user accounts are screened out, the user account with the smallest time difference between the recycling time and the purchase time of the user account is selected as the target user account; If no matching user account is found, the user account associated with the most recent purchase record is selected as the target user account based on the recycling time.

6. The method for recommending products for a self-service vending machine according to claim 1, wherein: The camera of the self-service vending machine includes a main camera and an auxiliary camera. If the cabinet door is detected to be open, the step of determining the user's pickup of the goods based on the video data collected by the camera includes: Detecting that the cabinet door is opened, obtaining an initial state image of the interior of the self-service vending machine captured by the main camera, and a hand movement trajectory during the cabinet door opening period captured by the auxiliary camera; Detecting that the cabinet door is closed, obtaining a final state image of the interior of the self-service vending machine captured by the main camera; Comparing the final state image with the initial state image to determine the commodity location where the commodity state has changed; Determining the hand's stopping position in front of each of the commodity shelves based on the hand's movement trajectory; The product to be picked up by the user is determined based on the matching relationship between the product location and the stop position.

7. The method for recommending products for a self-service vending machine according to claim 6, wherein: The step of determining the hand's stopping position in front of each of the commodity storage locations based on the hand's movement trajectory includes: Dividing the hand movement trajectory into a plurality of sub-trajectory segments according to a preset time window, wherein the sub-trajectory segments correspond to the continuous movement paths of the hand within the commodity storage location; Determining, based on the starting coordinates and the ending coordinates of the sub-trajectory segment, the entry time point of the hand into the commodity storage location and the exit time point of the hand leaving the commodity storage location; Determining a stay time of the hand at the corresponding commodity location according to the entry time point and the exit time point; The dwell time is compared with a preset dwell time threshold to determine the dwell position of the hand in front of each of the commodity shelves.

8. The method for recommending products for a self-service vending machine according to claim 1, wherein: The step of obtaining keywords corresponding to the pickup products and generating a self-mixed beverage search command based on the keywords includes: Query the pre-stored product keyword library to obtain the first keyword corresponding to the pickup product; According to the additional features of the pickup product, a second keyword corresponding to the pickup product is obtained; The first keyword and the second keyword are combined according to a preset template to generate the self-mixed beverage search command.

9. A self-service vending machine, characterized in that: The self-service vending machine device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the self-service vending machine commodity recommendation method according to any one of claims 1 to 8.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the self-service vending machine product recommendation method according to any one of claims 1 to 8 are implemented.

Images