A method and system for updating a store of a medical supply vending machine

Abstract

The application discloses a mall updating method and system for a medical product vending machine, detects a commodity loading signal, controls image scanning devices of all commodity channels to scan and analyze commodities in each commodity channel, and updates commodity conditions in each commodity channel; according to the commodity conditions in each commodity channel, corresponding mark types are marked for each commodity channel; meanwhile, a commodity information updating request is sent to a mall control terminal; the mall control terminal responds to the commodity information updating request, reads mark types of each commodity channel in sequence, and when the mark type is a target type, acquires a package scanning image on a surface of all commodities in the commodity channel; AI large models are used to search for corresponding commodity information of the package scanning image, generate commodity updating information of the commodity channel, and automatically update commodity display information of the commodity channel in a mall client. Commodity information display contents of the commodity channel in the mall client are automatically updated through the commodity channel type.

Description

Technical Field

[0001] This invention relates to the field of medical device retail technology, specifically to a method and system for updating the online store of a medical supplies vending machine. Background Technology

[0002] To make it easier for patients to quickly purchase the medical supplies they need without having to queue at the pharmacy, unmanned medical supply vending machines are now placed in every department and ward of the hospital. Patients can simply log in to the online store to place an order or scan the QR code of the mini-program to purchase the medical supplies they want from the unmanned medical supply vending machine.

[0003] Currently, unmanned medical supply vending machines on the market use the traditional manual restocking model. This involves staff manually placing similar medical supplies from the same batch into designated aisles according to the hospital's bulk purchase order, and then manually updating the product information in the backend e-commerce client. However, staff need to restock and check vending machines in various hospital floors, departments, and wards, resulting in a heavy workload. It's easy for staff to forget to update the e-commerce client information promptly after restocking, leading to a situation where vending machine products are updated, but the e-commerce client's product information is not updated in a timely manner.

[0004] Therefore, how to automatically update the product information on the e-commerce client while updating the vending machine's products is an urgent problem to be solved. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for updating the online store of medical supplies vending machines. By automatically identifying the product update information in the vending machine, the system synchronously updates the relevant parts of the product update in the display interface of the backend online store, thereby automatically updating the product information display content of the online store client while the vending machine updates its products.

[0006] To achieve the above objectives, this application provides the following solution:

[0007] On one hand, this invention provides a method for updating an online store for a medical supplies vending machine, comprising an interactive vending machine terminal, an online store control terminal, and a product client, the specific interaction process of which is as follows:

[0008] Vending machine terminal: Upon detecting a product loading signal, it controls the image scanning devices of all product lanes to scan the products in each lane;

[0009] When the vending machine detects a signal that the product refill is complete, it analyzes the product update status of each product lane based on the product scanning results of each lane, and marks each lane with the corresponding tag type based on the product update status of each lane;

[0010] Send a commodity information update request to the mall control terminal at the same time;

[0011] Mall control terminal: in response to the commodity information update request, read the marker type of each aisle in turn, and obtain the packaging scan image on the surface of all commodities in the aisle when the marker type is the target type;

[0012] According to the packaging scan image on the surface of all commodities in the aisle, search for the corresponding commodity information of the packaging scan image using an AI large model, and generate commodity update information for the aisle;

[0013] Send the commodity update information of the aisle to the mall client;

[0014] Mall client: receiving the commodity update information, automatically updating the commodity display information of the aisle in the mall client according to the commodity update information corresponding to the aisle.

[0015] In some specific embodiments, the specific process of marking each aisle with the corresponding marker type according to the commodity update of each aisle is as follows:

[0016] S01, generate a commodity loading signal when the vending machine door opening action is monitored, and control the image scanning device of all aisles to record videos of the aisles respectively, stop recording when the vending machine door signal is monitored, and obtain the recorded video segment;

[0017] S02, analyze the recorded video segment to analyze whether the number of commodities in each aisle changes;

[0018] S03, according to the trend of the change of the number of commodities, mark the aisle as a new aisle or a replenishment aisle, and mark the aisle with no change in the trend of the number of commodities as a non-new aisle.

[0019] In some specific embodiments, the specific process of step S03 is as follows:

[0020] Read a number of frames of images from the recorded video segment, and store the number of frames of images in the analysis sequence in turn,

[0021] Obtain the first and last frames of images from the analysis sequence, and compare the change of the number of commodities in the first and last frames of images;

[0022] If the number of commodities in the first and last frames of images changes, identify the number of commodities in each frame of image in the analysis sequence in turn, analyze the trend of the change of the number of commodities, and mark the aisle as a replenishment aisle when the trend of the change of the number of commodities is linearly increasing;

[0023] When the trend of the change of the number of commodities is first decreasing and then increasing, mark the aisle as a new aisle.

[0024] In some specific embodiments, the replenishment aisle includes a full replenishment aisle and a replenishment new product aisle, and the specific process is as follows:

[0025] When the change trend of the number of goods is gradually changed from zero to the target number, the aisle is marked as a full replenishment aisle;

[0026] When the change trend of the number of goods is gradually increased from n to the target number m, and m>n, the aisle is marked as a replenishment new product aisle.

[0027] In some specific embodiments, if the number of goods in the first and last frames of images does not change, any intermediate frame of images is obtained;

[0028] If the number of goods in the first and last frames of images is equal to the number of goods in any intermediate frame of images, the aisle is marked as a non-new product aisle;

[0029] If the number of goods in the first and last frames of images is equal and different from the number of goods in any intermediate frame of images, the aisle is marked as a new product aisle.

[0030] In some specific embodiments, when the target type is a new product aisle, the process of generating the goods update information of the aisle is as follows:

[0031] Obtain the last frame of images of the recorded video segment, divide the single product upper surface packaging scan graph corresponding to each product from the last frame of images, divide all single product upper surface packaging scan graphs according to the packaging face view of the product, and count the number of goods;

[0032] Input each packaging face view of the product into an AI large model for image recognition, search for the same product information as the packaging face view of the product, the product information including the complete six-face packaging view of the product, the product name, and the product manufacturer, and write the product information and the number of goods into the goods update information.

[0033] In some specific embodiments, when the target type is a full replenishment aisle, the process of generating the goods update information of the aisle is as follows:

[0034] Obtain the last frame of images of the recorded video segment, divide the single product upper surface packaging scan graph corresponding to each product from the last frame of images, and count the number of goods;

[0035] Call the product six-view of the historical product display information of the aisle in the mall client, query whether the single product upper surface packaging scan graph exists in the product six-view of the historical product information, if it exists, the goods update information only includes the number of goods update;

[0036] If it does not exist, divide all single product upper surface packaging scan graphs according to the packaging face view of the product;

[0037] The various package views of the commodity are input into the AI large model for image recognition, and the same commodity information as the package view of the commodity is searched, the commodity information including the complete six-package views of the commodity, the commodity name, and the commodity manufacturer, and the commodity information and the quantity of the commodity are written into the commodity update information.

[0038] In some specific embodiments, when the target type is a replenishment new goods channel, the process of generating the commodity update information of the goods channel is:

[0039] The first frame image and the last frame image of the recorded video segment are obtained, the last frame image is subtracted from the first frame image to obtain a difference image, and the single commodity upper surface package scanning image corresponding to each commodity is divided from the first frame image and the difference image respectively, and the number of commodities is counted to obtain the first commodity image set corresponding to the first frame image and the first commodity number, the second commodity image set corresponding to the difference image and the second commodity number;

[0040] According to the first commodity image set and the second commodity image set, it is judged whether the first commodity and the second commodity are the same, if the first commodity and the second commodity are the same, the commodity update information only includes the commodity update number, and the commodity update number is the sum of the first commodity number and the second commodity number;

[0041] If the first commodity and the second commodity are not the same, the second commodity image set is input into the AI large model for image recognition, and the same commodity information as the package view of the commodity is searched, the commodity information including the complete six-package views of the commodity, the commodity name, and the commodity manufacturer, and the commodity information and the second commodity quantity are written into the commodity update information.

[0042] In some specific embodiments, if the first commodity and the second commodity are not the same, the commodity update monitoring program is called to monitor the change of the first commodity quantity in the replenishment new goods channel of the vending machine terminal, and when the first commodity quantity is reduced to 0, the commodity update information of the replenishment new goods channel is sent to the mall client.

[0043] In a second aspect, the application provides a mall update system for a medical supply vending machine, comprising: a vending machine terminal, a mall control terminal and a commodity client which interact with each other, wherein:

[0044] The vending machine terminal comprises a vending machine body and a control module, the vending machine body is provided with a plurality of shelves, each shelf comprises a plurality of goods channels, an image scanning device is installed directly above each goods channel, the image scanning device is used for scanning the image directly above the goods in the goods channel, the control module is used for controlling all the image scanning devices of the goods channels to scan the goods in each goods channel when a goods loading signal is detected, analyzing the goods update of each goods channel according to the scanning results of the goods in each goods channel when a goods loading completion signal is detected, marking each goods channel with a corresponding mark type according to the goods update of each goods channel, and sending a goods information update request to a mall control terminal;

[0045] The mall control terminal is used for responding to the goods information update request, reading the mark types of the goods channels in sequence, and acquiring the package scanning diagram of the upper surfaces of all the goods in the goods channel when the mark type is a target type;

[0046] According to the package scanning diagram of the upper surfaces of all the goods in the goods channel, the AI large model is used to search the corresponding goods information of the package scanning diagram, and the goods update information of the goods channel is generated;

[0047] The goods update information of the goods channel is sent to a mall client;

[0048] The mall client is used for automatically updating the goods display information of the goods channel in the mall client according to the received goods update information of the goods channel.

[0049] The application has the beneficial effects that:

[0050] The application monitors the opening and closing door signals of the vending machine, sends a vending machine goods loading completion signal when a closing door signal is detected, controls the image scanning device to scan the goods in each goods channel at this time, identifies the goods images scanned, obtains the new goods information of the goods channel, marks each goods channel according to the goods quantity change, judges whether the goods display interface in the mall client corresponding to the goods channel needs to be updated according to the mark type of the goods channel, updates the goods information in the mall client corresponding to the goods channel according to the new goods information if the goods display interface needs to be updated, does not need the goods loading personnel to manually input the goods update information, automatically updates the display page content uploaded in the background at the same time of the vending machine goods update information, provides the latest information for the patient in time, and helps the patient to select the correct goods. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 A mall update method flow chart for a medical product vending machine provided for the embodiment of the application;

[0052] Figure 2 A goods channel mark logic schematic diagram provided for the embodiment of the application;

[0053] Figure 3 A replenishment and new product channel commodity update information generation logic schematic diagram provided for an embodiment of the present application is shown in FIG. 1.

[0054] Figure 4 A commodity display page schematic diagram provided for an embodiment of the present application is shown in FIG. 2. DETAILED DESCRIPTION

[0055] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. The description of the at least one exemplary embodiment is actually only illustrative, but not as any limitation on the present application and its application or use. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort belong to the scope of protection of the present application.

[0056] The relative arrangement, numerical expressions and values of the components and steps set forth in the embodiments do not limit the scope of the present application, unless otherwise specifically stated.

[0057] It should be understood that the sizes of the various parts shown in the drawings are not drawn in accordance with the actual proportional relationship for the convenience of description.

[0058] In addition, the description of well-known structures, functions and configurations can be omitted for the sake of clarity and conciseness. Those of ordinary skill in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of the present disclosure.

[0059] The technology, methods and devices known to those of ordinary skill in the relevant art can not be discussed in detail, but should be considered as part of the authorized specification under appropriate circumstances.

[0060] In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary, and not as a limitation. Thus, other examples of the exemplary embodiments can have different values.

[0061] Embodiment 1

[0062] As shown in FIG. 1, the present embodiment provides a mall update method for a medical product vending machine, which includes a vending machine terminal, a mall control terminal and a commodity client that interact with each other, and the specific process of the interaction is as follows: Figure 1 Vending machine terminal: detects a commodity restocking signal to control the image scanning device of all channels to scan the commodities of each channel.

[0063]

[0064] ​The vending machine detects the completion of the goods loading signal, analyzes the goods update of each channel according to the goods scanning result of each channel, and marks each channel according to the goods update of each channel;

[0065] Meanwhile, the vending machine sends a goods information update request to the mall control terminal;

[0066] The mall control terminal: in response to the goods information update request, reads the mark type of each channel in turn, and when the mark type is the target type, obtains the package scanning image on the surface of all goods in the channel;

[0067] According to the package scanning image on the surface of all goods in the channel, the AI large model searches for the corresponding goods information of the package scanning image, and generates the goods update information of the channel;

[0068] The goods update information of the channel is sent to the mall client;

[0069] The mall client: receiving the goods update information, automatically updating the goods display information of the channel in the mall client according to the goods update information corresponding to the channel.

[0070] It can be understood that an image scanning device is installed above each channel in the vending machine, which is used to scan the image above the goods in the channel. In order to reduce the recognition pressure of the background, the goods update of the channel in the vending machine is judged in advance, and the channel in the vending machine is marked. According to the mark type of the channel, it is judged whether it needs to be updated in the mall client. When the vending machine door signal is received, the vending machine goods loading completion signal is generated at this time, and the goods update request is sent to the mall control center to further analyze whether the goods display page of the mall client needs to be updated.

[0071] S1, receiving the vending machine goods loading completion signal, obtaining the mark type of each channel, the mark type including new goods channel and replenishment goods channel;

[0072] Specifically, the specific process of marking each channel in advance in the vending machine terminal is as follows:

[0073] S01, when the vending machine door opening action is monitored, a goods loading signal is generated, and the image scanning device of all channels is controlled to record the video of the channel respectively. When the vending machine door signal is monitored, the image scanning device is controlled to stop recording, and the recording video segment is obtained;

[0074] Generally, when the stocker stocks, the stocker will open the cabinet door, and then fill the goods on the new goods channel or the goods channel that needs to be replenished. After filling the vending cabinet, the cabinet door is closed. Therefore, a sensor can be arranged at the cabinet door to detect the opening and closing signals of the vending cabinet. When the opening signal is detected, the goods stocking signal is triggered to start video recording of each goods channel of the vending cabinet for subsequent analysis.

[0075] S02, analyzing the recorded video segment to analyze whether the number of goods in each goods channel changes;

[0076] S03, according to the trend of the change of the number of goods, marking the goods channel as a new goods channel or a replenishment goods channel, and marking the goods channel whose trend of change of the number of goods does not change as a non-new goods channel.

[0077] As shown in Figure 2 , the specific process of marking each goods channel is as follows:

[0078] S031, reading a plurality of frames of images from the recorded video segment by frame skipping, and sequentially storing the plurality of frames of images in an analysis sequence,

[0079] S032, obtaining the first and last frames of images from the analysis sequence, and comparing the change of the number of goods in the first and last frames of images;

[0080] SA: if the number of goods in the first and last frames of images does not change, obtaining any frame of image between the first and last frames of images;

[0081] If the number of goods in the first and last frames of images is equal to the number of goods in any frame of image between the first and last frames of images, the goods channel is marked as a non-new goods channel;

[0082] If the number of goods in the first and last frames of images is equal and different from the number of goods in any frame of image between the first and last frames of images, the goods channel is marked as a new goods channel.

[0083] SB: if the number of goods in the first and last frames of images changes, sequentially identifying the number of goods in each frame of image in the analysis sequence, and analyzing the trend of the change of the number of goods. When the trend of the change of the number of goods is linearly increasing, the goods channel is marked as a replenishment goods channel;

[0084] When the trend of the change of the number of goods is first decreasing and then increasing, the goods channel is marked as a new goods channel.

[0085] The replenishment goods channel includes a full replenishment goods channel and a replenishment new goods channel, and the specific process is as follows:

[0086] When the trend of the change of the number of goods is gradually changing from zero to the target number, the goods channel is marked as a full replenishment goods channel;

[0087] When the commodity quantity change trend is gradually increasing from n to target quantity m, and m>n, mark the goods shelf as replenishment new goods shelf.

[0088] It can be understood that in order to determine the change of the commodity quantity of the goods shelf, the video segment can be analyzed, the first and last frames of images can be analyzed to obtain the change of the commodity of the goods shelf, then the video segment is analyzed according to the change of the commodity of the goods shelf to obtain images by frame skipping and analyze the images frame by frame, the images obtained by frame skipping are screened, only the images containing only commodities and the adjacent two frames of images have changed (this is to avoid the case that the hands of the goods personnel block the commodities, and these images are excluded), then the commodity contour information in the screened images is recognized, that is, the commodity quantity included in each frame of image is recognized, and then the change of the commodity quantity of the goods shelf is analyzed according to the images.

[0089] The commodities of each goods shelf are analyzed, if the two frames of images of the start frame and the end frame (corresponding to the first and last frames of images) are completely the same and have not changed, it indicates that the goods shelf may be in a full load condition and does not need to be replenished or supplemented, or the goods shelf needs to be completely replaced with new commodities, that is, the commodities in the goods shelf are all taken off and new commodities are loaded, therefore, it is necessary to further distinguish the above two cases by combining any one of the images between the first and last frames of images, when the quantities are equal, the goods shelf and the corresponding commodities are skipped at this time, the goods shelf is marked as a new goods shelf, and subsequent operations are not needed and the commodity information displayed in the mall client of the goods shelf does not need to be updated. If the goods shelf needs to be completely replaced with new commodities, it is considered that the goods shelf has been replaced with new commodities, the commodity interface needs to be updated, and the goods shelf is marked as a new goods shelf.

[0090] If the two frames of images of the start frame and the end frame have changed, it indicates that the goods shelf has been replenished or supplemented, and the two cases need to be distinguished, if the commodities are new, the commodity page display information in the mall client needs to be updated. If the commodities are updated, even if the goods shelf still has a small amount of commodities, but the commodity quantity of the goods shelf presents a trend of decreasing from a small amount to 0 and then increasing from 0 to a full load, at this time, it is considered that the goods shelf has been replenished with new commodities, the goods shelf is marked as a new goods shelf, and it is indicated that the commodity display page corresponding to the commodity in the mall display page needs to be updated.

[0091] If the commodities are supplemented, the commodity quantity presents a trend of increasing, at this time, two cases can occur, the first case is that the commodities in the goods shelf are all sold out and need to be refilled, that is, the commodities need to be supplemented, the goods shelf can be marked as a full replenishment goods shelf, and after receiving the closing signal, the goods shelf is returned to the mall control terminal for further judgment.

[0092] The second is that the goods channel has not been sold out, leaving a small amount of goods, and the goods supplier only needs to use the goods to fill the goods channel. At this time, two situations will occur. One is that the goods supplier uses the same kind of goods to complete the filling of the goods channel, and the other is to use new goods (which may have the same name but different packaging, or may be two completely different goods) to complete the filling of the goods channel. At this time, observing the change in the number of goods, the number of goods will change from few to many and eventually be full, indicating that the goods channel belongs to replenishment, and the goods channel is marked as a replenishment new goods channel. However, whether the replenishment goods are original goods or new goods needs to be further judged in the control terminal of the mall.

[0093] The change of goods in each goods channel has been analyzed above, and the update of goods in the goods channel has been preliminarily judged, and based on this, each goods channel is marked with a corresponding mark type. When the vending machine detects the goods information update request sent by the vending machine to the mall control terminal, the mark type of each goods channel and the recorded video segment are sent to the mall control terminal, and the mall control terminal further generates corresponding goods update information based on the goods channel type and sends it to the mall client for update. For different goods channels, the goods update information is also different. Since the goods supplier places the goods randomly when placing the goods, reading only the upper surface packaging information of the goods may not be able to identify the goods information. Generally, a goods channel is a batch of the same and similar goods, and it is also possible that the goods are similar but have different packaging. Therefore, the upper surface image of all goods in the goods channel can be comprehensively judged, and then the remaining information can be filled by using an AI model to generate goods information and encapsulate it into update information and send it to the mall client to update the goods display page. As shown in Figure 4 The goods update information includes basic information and detailed information. The basic information includes the goods channel number corresponding to the goods channel, the vending machine number, and the area where the vending machine is located (the area where the vending machine is located, the department, etc.). The detailed information includes goods information and the number of goods. The goods information includes the complete six-packaging view of the goods (the complete six-packaging view is divided into the main picture and the detail picture of the goods display page by the mall client), the name of the goods, the manufacturer of the goods, and the selling price of the goods. For different mark types of goods channels, the mall updates different information, which can be divided into the following types:

[0094] 1. When the target type is a new goods channel, the process of generating goods update information for the goods channel is as follows:

[0095] Get the last frame image of the recorded video segment, and divide the packaging scan image of the upper surface of each single goods from the last frame image. Divide all single goods upper surface packaging scan images according to the packaging view of the goods and count the number of goods.

[0096] Input each package view of the commodity into the AI large model for image recognition, search the same commodity information as the package view of the commodity, the commodity information includes the complete six-package view of the commodity, the commodity name, and the commodity manufacturer, write the commodity information and the commodity quantity into the commodity update information.

[0097] 2. When the target type is a full replenishment aisle, the process of generating the commodity update information of the aisle is:

[0098] Obtain the last frame image of the recorded video segment, divide the package scanning image of the upper surface of each single commodity corresponding to the commodity from the last frame image, and count the commodity quantity;

[0099] Call the six-view of the commodity in the historical commodity display information of the aisle in the mall client, query whether the package scanning image of the upper surface of the single commodity exists in the six-view of the historical commodity information, if exists, the commodity update information only includes the commodity quantity update;

[0100] If not, divide all the package scanning images of the upper surface of the single commodity according to the package view of the commodity;

[0101] Input each package view of the commodity into the AI large model for image recognition, search the same commodity information as the package view of the commodity, the commodity information includes the complete six-package view of the commodity, the commodity name, and the commodity manufacturer, write the commodity information and the commodity quantity into the commodity update information.

[0102] 3. When the target type is a replenishment and new commodity aisle, as shown in Figure 3 the process of generating the commodity update information of the aisle is:

[0103] Obtain the first frame image and the last frame image of the recorded video segment, difference the last frame image and the first frame image to obtain a difference image, divide the package scanning image of the upper surface of each single commodity corresponding to the commodity from the first frame image and the difference image respectively and count the commodity quantity, obtain the first commodity image set corresponding to the first frame image and the first commodity quantity, the second commodity image set corresponding to the difference image and the second commodity quantity;

[0104] According to the first commodity image set and the second commodity image set, determine whether the first commodity and the second commodity are the same, if the first commodity and the second commodity are the same, the commodity update information only includes the commodity update quantity, and the commodity update quantity is the sum of the first commodity quantity and the second commodity quantity;

[0105] If the first commodity and the second commodity are not the same, the second commodity image set is input into the AI large model for image recognition, and the same commodity information as the package surface view of the commodity is searched, the commodity information includes the complete six-package view of the commodity, the commodity name, and the commodity manufacturer, and the commodity information and the second commodity quantity are written into the commodity update information. If the first commodity and the second commodity are not the same, the commodity update monitoring program is called to monitor the change of the first commodity quantity in the new commodity channel of the vending machine terminal, and when the first commodity quantity is reduced to 0, the commodity update information of the new commodity channel is sent to the mall client.

[0106] It can be seen that when the commodity update information includes the commodity update quantity, it means that the commodity channel is restocked at this time, and the commodity types before and after restocking do not change, so only the display quantity of the mall client needs to be updated. When the commodity update information includes the commodity information and the commodity update quantity, it means that the commodity has changed, so not only the commodity quantity needs to be updated, but also the commodity image of the commodity display page of the mall client needs to be replaced to match the text and image. When the commodity type is replaced, there is another case, that is, when there are two different packaged commodities in the commodity channel, it may be a new commodity and a commodity corresponding to the commodity display information. At this time, the change of the commodity quantity can be monitored (usually the latest commodity is placed at the back by the restocking personnel, that is, the commodity channel sells the remaining commodities first and then sells the newly restocked commodities), and when it is monitored that the commodity channel has sold out the original commodities, the update program is automatically called to update the commodity display page corresponding to the commodity channel in the display page content of the mall client.

[0107] For the mall client, when receiving the commodity update information, the commodity update information is parsed, and the display page corresponding to the commodity channel is located according to the basic information in the commodity update information. If the detailed information of the commodity update information only includes the commodity quantity, only the commodity quantity of the commodity in the display page of the commodity channel is updated. When the detailed information includes the commodity information, the commodity display image and the commodity information of the display page are replaced. Specifically, it can be understood that the part of the commodity information to be replaced in the commodity display page includes a plurality of labels to be filled, and the label types to be filled include text or image.

[0108] The part of the commodity information to be replaced in the commodity display page is replaced, when the label type in the replacement part is text and the label name is the same as the field name corresponding to the commodity information, the corresponding field value in the commodity information is filled into the text label to be filled;

[0109] When the label type in the replacement part is image and the label name is the same as the field name corresponding to the commodity information, the corresponding image in the commodity information is filled into the image label to be filled.

[0110] Embodiment 2

[0111] The embodiment is used to realize the mall updating method of embodiment 1, and provides a mall updating system for a medical supply vending machine, which comprises vending machine terminals, a mall control terminal and a commodity client which interact with each other, wherein:

[0112] The vending machine terminal comprises a vending machine body and a control module. The vending machine body is provided with a plurality of shelves, each shelf comprises a plurality of channels, and an image scanning device is installed directly above each channel. The image scanning device is used for scanning the image directly above the commodity in the channel. The control module is used for detecting a commodity loading signal to control the image scanning devices of all channels to scan the commodities in the channels, detecting a commodity loading completion signal, analyzing the commodity updating conditions of each channel according to the scanning results of the commodities in the channels, marking each channel with a corresponding mark type according to the commodity updating conditions of each channel, and sending a commodity information updating request to the mall control terminal;

[0113] The mall control terminal is used for responding to the commodity information updating request, reading the mark types of the channels in sequence, and obtaining the package scanning images of all commodities on the surfaces of the channels when the mark type is a target type;

[0114] According to the package scanning images of all commodities on the surfaces of the channels, the AI large model is used to search for the commodity information corresponding to the package scanning images, and commodity updating information of the channels is generated;

[0115] The commodity updating information of the channels is sent to the mall client;

[0116] The mall client is used for automatically updating the commodity display information of the channels in the mall client according to the received commodity updating information of the channels.

[0117] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. According to the technical essence of the present application, any simple modification, equivalent replacement and improvement of the above embodiment within the spirit and principles of the present application are still within the protection scope of the technical solution of the present application.

Claims

1. A method for updating a store of a medical supply vending machine, characterized by, The vending machine terminal, the mall control terminal and the commodity client are interacted with each other, and the specific process of the interaction is as follows: The vending machine terminal: detecting the commodity loading signal controls the image scanning device of all the goods channels to scan the goods in each goods channel; When the vending machine detects the commodity loading completion signal, the commodity update situation of each goods channel is analyzed according to the scanning results of the goods in each goods channel, and the corresponding mark type is marked for each goods channel according to the commodity update situation of each goods channel; the mark type includes new goods channel, full restocking goods channel and restocking new goods channel; At the same time, the commodity information update request is sent to the mall control terminal; The mall control terminal: in response to the commodity information update request, the mark type of each goods channel is read in turn, and when the mark type is the target type, the package scanning graph on the surface of all the goods in the goods channel is obtained; According to the package scanning graph on the surface of all the goods in the goods channel, the AI large model is used to search the corresponding commodity information of the package scanning graph, and the commodity update information of the goods channel is generated; When the target type is the new goods channel, the process of generating the commodity update information of the goods channel is as follows: The last frame image of the recorded video segment is obtained, the package scanning graph of the single upper surface of each commodity corresponding to the last frame image is divided, and the number of commodities is counted according to the package surface view of the commodity; The package surface view of the commodity is input into the AI large model for image recognition, the same commodity information as the package surface view of the commodity is searched, the commodity information includes the complete six-face package view of the commodity, the name of the commodity and the manufacturer of the commodity, and the commodity information and the number of commodities are written into the commodity update information; When the target type is the full restocking goods channel, the process of generating the commodity update information of the goods channel is as follows: The last frame image of the recorded video segment is obtained, the package scanning graph of the single upper surface of each commodity corresponding to the last frame image is divided, and the number of commodities is counted; The complete six-face package view of the historical commodity display information of the goods channel in the mall client is called, and it is inquired whether the package scanning graph of the single upper surface of the commodity exists in the complete six-face package view of the historical commodity information. If it exists, the commodity update information only includes the number of commodity updates; If it does not exist, all the package scanning graphs of the single upper surface of the commodity are divided according to the package surface view of the commodity; The package surface view of the commodity is input into the AI large model for image recognition, the same commodity information as the package surface view of the commodity is searched, the commodity information includes the complete six-face package view of the commodity, the name of the commodity and the manufacturer of the commodity, and the commodity information and the number of commodities are written into the commodity update information; When the target type is the restocking new goods channel, the process of generating the commodity update information of the goods channel is as follows: The first frame image and the last frame image of the recorded video segment are obtained, the last frame image is subtracted from the first frame image to obtain a difference image, the package scanning graph of the single upper surface of each commodity corresponding to the first frame image and the difference image is divided, and the number of commodities is counted to obtain the first commodity image set corresponding to the first frame image and the first commodity number, the second commodity image set corresponding to the difference image and the second commodity number. According to the first commodity image set and the second commodity image set, it is judged whether the first commodity and the second commodity are the same, and if the first commodity and the second commodity are the same, only the commodity update quantity is included in the commodity update information, and the commodity update quantity is the sum of the first commodity quantity and the second commodity quantity; If the first commodity and the second commodity are not the same, the second commodity image set is input into the AI large model for image recognition, the same commodity information as the package surface view of the second commodity is searched, the commodity information includes the complete six-package view of the commodity, the commodity name and the commodity manufacturer, and the commodity information and the second commodity quantity are written into the commodity update information; The commodity update information of the goods channel is sent to the mall client; The mall client: receiving the commodity update information, automatically updating the commodity display information of the goods channel in the mall client according to the goods channel corresponding to the commodity update information.

2. The method of claim 1, wherein, The specific process of marking each goods channel with a corresponding mark type according to the commodity update situation of each goods channel is: S01, when the vending machine door opening action is monitored, a commodity restocking signal is generated, and the image scanning device of all goods channels is controlled to record videos of the goods channels respectively, and when the vending machine door signal is monitored, the image scanning device is controlled to stop recording, and a recorded video segment is obtained; S02, analyzing the recorded video segment, analyzing whether the commodity quantity of each goods channel changes; S03, according to the trend of commodity quantity change, mark the goods channel as a new goods channel or a replenishment goods channel, and mark the goods channel with no change in commodity quantity change trend as a non-new goods channel.

3. The method of claim 2, wherein, The specific process of S03 is: Jump frame reading of several frames of images from the recorded video segment, and sequentially storing the several frames of images into an analysis sequence, Obtain the first and last two frames of images from the analysis sequence, and compare the change of the commodity quantity in the first and last two frames of images; If the commodity quantity in the first and last two frames of images changes, then identify the commodity quantity of each frame of image in the analysis sequence in turn, analyze the commodity quantity change trend, and when the commodity quantity change trend is linearly increasing, mark the goods channel as a replenishment goods channel; When the commodity quantity change trend is first decreasing and then increasing, mark the goods channel as a new goods channel.

4. The method of claim 3, wherein, The replenishment goods channel includes a full replenishment goods channel and a replenishment new goods channel, and the specific process is: When the commodity quantity change trend is from zero to the target quantity, mark the goods channel as a full replenishment goods channel; When the commodity quantity change trend is from n to gradually increase to the target quantity m, and m>n, mark the goods channel as a replenishment new goods channel.

5. The method of claim 3, wherein, If the commodity quantity in the first and last two frames of images does not change, then obtain any one frame of image in the middle of the first and last two frames of images; If the commodity quantity of the first and last two frames of images is equal to the commodity quantity of any one frame of image in the middle, mark the goods channel as a non-new goods channel; If the commodity quantity of the first and last two frames of images is equal and different from the commodity quantity of any one frame of image in the middle, mark the goods channel as a new goods channel.

6. The method of claim 1, wherein, If the first commodity and the second commodity are not the same, call the commodity update monitoring program to monitor the first commodity quantity change of the replenishment new goods channel of the vending machine terminal, and when the first commodity quantity decreases to 0, send the commodity update information of the replenishment new goods channel to the mall client.

7. A mall updating system for a medical supply vending machine, which applies a mall updating method for a medical supply vending machine as claimed in claim 1, characterized by, It includes: The vending machine terminal, the mall control terminal and the commodity client interact with each other, wherein: The vending machine terminal comprises a vending machine body and a control module, the vending machine body is provided with a plurality of layers of shelves, each layer of shelves comprises a plurality of goods channels, an image scanning device is installed directly above each goods channel, the image scanning device is used for scanning the image directly above the goods in the goods channel, the control module is used for controlling all the image scanning devices of the goods channels to scan the goods in each goods channel when a goods loading signal is detected, analyzing the updating situation of the goods in each goods channel according to the scanning results of the goods in each goods channel when a goods loading completion signal is detected, marking each goods channel with a corresponding mark type according to the updating situation of the goods in each goods channel, and sending a goods information updating request to a mall control terminal; The mall control terminal is used for responding to the goods information updating request, reading the mark types of the goods channels in sequence, and acquiring the package scanning diagram of the upper surfaces of all the goods in the goods channel when the mark type is a target type; According to the package scanning diagram of the upper surfaces of all the goods in the goods channel, a goods information corresponding to the package scanning diagram is searched by using an AI large model, and goods updating information of the goods channel is generated; The goods updating information of the goods channel is sent to a mall client; The mall client is used for automatically updating the goods display information of the goods channel in the mall client according to the received goods updating information of the goods channel.

Images