A query and monitoring system for automated drug vending machines

Abstract

This invention provides a query and monitoring system for automated drug vending machines, comprising: an identification and traceability terminal for authenticating the source of purchased drugs based on their source identification; an inventory recording terminal for registering batch numbers of all source-authenticated qualified drugs among the purchased drugs and obtaining current inventory information and purchase records for each qualified drug; a full-process monitoring terminal for monitoring the qualified drugs throughout the entire process and obtaining corresponding monitoring information; and an information update terminal for updating the current inventory information based on the monitoring information and generating corresponding drug circulation records. This system is used for verifying the legality of drug purchases and automatically entering information, as well as monitoring and managing drug storage. It achieves real-time and accurate management of drug purchase sources, sales, inventory, and expiration dates and batch numbers, and records detailed information for each sale, enabling transparent traceability of drug circulation and inventory management by pharmacies.

Description

Technical Field

[0001] This invention relates to the field of vending machine technology, and in particular to a query and monitoring system for a pharmaceutical vending machine. Background Technology

[0002] Currently, with the development of technology, consumers are finding more diverse ways to purchase goods, leading to the emergence of various vending machines, including automatic medicine vending machines. Driven by the new retail model, many pharmacies are installing these machines, allowing consumers to experience the benefits of "Internet + healthcare" innovation firsthand. Equipping medical institutions' outpatient halls with automatic medicine vending machines not only adapts to the changing demands of drug consumption but also leverages the advantages of their network to expand convenient value-added services. During the pandemic, these machines can also reduce patient waiting times, enabling contactless medication purchases and preventing cross-infection.

[0003] Automated medicine vending machines differ significantly from ordinary vending machines. Firstly, medicines require specific cool or room temperature storage conditions, so vending machines must ensure these conditions comply with Good Manufacturing Practices (GMP) for pharmaceuticals to guarantee quality. Secondly, the pharmacy's back-end service center must accurately manage the purchase source, sales, inventory, and expiration dates of medicines in real time. It should also be able to monitor each vending machine in real time, record detailed sales information, and upload data in real-time to facilitate transparent traceability of medicine distribution and inventory management for the pharmacy. However, most existing automated medicine vending machines primarily function as medicine sellers, offering relatively limited functionality and failing to cover the aforementioned requirements.

[0004] Therefore, this invention proposes a query and monitoring system for automated drug vending machines. Summary of the Invention

[0005] This invention provides a query and monitoring system for automated drug vending machines. Based on the source identifier of purchased drugs, it verifies the source of the drugs and registers their batch numbers in the warehouse. This allows for obtaining the current inventory information of the vending machine and the purchase record of each qualified drug, enabling the verification of the legality of drug purchases and automatic information entry. Furthermore, by monitoring the qualified drugs throughout the entire process, it achieves monitoring and management of drug storage. The system also updates the current inventory information in real time and generates corresponding drug circulation records, enabling real-time and accurate management of drug purchase sources, sales, inventory, and expiration dates and batch numbers. It can monitor each automated drug vending machine in real time, recording detailed information for each sale, achieving transparent traceability of drug circulation and enabling pharmacies to manage drug inventory.

[0006] This invention provides a query and monitoring system for automated drug vending machines, comprising:

[0007] The identification traceability terminal is used to authenticate the source of the purchased drugs based on the source identification of the purchased drugs;

[0008] The inventory entry record terminal is used to register the batch numbers of all qualified drugs that have passed source certification among all purchased drugs and to obtain the current inventory information of the drug vending machine and the purchase record of each qualified drug.

[0009] The end-to-end monitoring terminal is used to monitor the qualified drugs throughout the entire process and obtain corresponding monitoring information;

[0010] The information update terminal is used to update the current inventory information based on the monitoring information, and at the same time, generate corresponding drug circulation records based on the purchase records and the monitoring information.

[0011] Preferably, the identification traceability terminal includes:

[0012] The identification module is used to identify the source identification marks set on the purchased medicines and obtain the corresponding source information.

[0013] The information discrimination module is used to determine the legality of the source of the purchased drugs based on the source information;

[0014] The drug review module is used to determine that when the source of the purchased drug is determined to be legal, the corresponding purchased drug is a qualified drug that has passed source certification; otherwise, the corresponding purchased drug is a non-qualified drug that has not passed source certification.

[0015] The illegal alarm module is used to store the source information of the unqualified drugs when the purchased drugs are determined to be unqualified drugs that have not passed the source certification, and at the same time, issue a corresponding alarm signal.

[0016] Preferably, the identifier recognition module includes:

[0017] The image acquisition unit is used to acquire an appearance image of the outer packaging of the purchased medicine at each preset angle to obtain a corresponding set of appearance images.

[0018] A contour recognition unit is used to identify the contour of the source identifier corresponding to the source identifier set on the purchased medicine in the set of appearance images;

[0019] An information identification unit is used to determine the corresponding source identification area in the set of appearance images based on the source identification outline, and to identify the source information of the purchased medicine in the source identification area.

[0020] Preferably, the contour recognition unit includes:

[0021] The model building subunit is used to identify the appearance outline of the purchased medicine based on the appearance image, and to build a corresponding shape model of the purchased medicine based on the appearance outline.

[0022] The matrix determination subunit is used to determine the first point cloud data corresponding to the source identifier under the standard spatial shape, and to identify the corresponding second point cloud data based on the shape model of the purchased medicine, and to determine the corresponding first spatial shape transformation matrix based on the first point cloud data and the second point cloud data.

[0023] The point cloud transformation subunit is used to perform edge detection on the appearance image, identify the edge contour contained in the appearance image, determine the third point cloud data corresponding to the edge contour in the shape model of the purchased medicine based on the position of the edge contour in the corresponding appearance image, and determine the fourth point cloud data corresponding to the edge contour under the standard spatial shape based on the first spatial shape transformation matrix and the third point cloud data.

[0024] The feature extraction subunit is used to extract the shape features of the edge contour under the standard spatial shape based on the fourth point cloud data;

[0025] The contour sorting subunit is used to filter out the edge contours of the shape features that are consistent with the standard shape features corresponding to the source identifier in the appearance image, and use them as the corresponding source identifier local contours. According to the spatial relationship between the preset angles of the appearance images to which the source identifier local contours belong, the unit sorts and summarizes all the source identifier local contours in the appearance image set to obtain the corresponding source identifier local contour set.

[0026] A standard processing subunit is used to determine the light and shadow gradient features in the appearance image to which the local contour of the source identifier belongs, and at the same time, determine the visual effect evaluation value of the local contour of the source identifier. The light and shadow gradient features in the appearance image to which the local contour of the source identifier belongs corresponding to the maximum visual effect evaluation value are used as standard light and shadow gradient features. Based on the standard light and shadow gradient features, the remaining appearance images in the appearance image set other than the standard appearance images corresponding to the standard light and shadow gradient features are subjected to unified light and shadow feature processing to obtain the corresponding standard appearance image set.

[0027] The contour splicing subunit is used to determine the edges to be spliced ​​in all local contours of standard source identifiers contained in the standard appearance image set, determine the local splicing discrimination edges in the corresponding local contours of standard source identifiers based on the edges to be spliced, extract grayscale gradient features from the local splicing discrimination edges to obtain the corresponding local grayscale gradient features, and overlap and connect the local splicing discrimination edges with consistent local grayscale gradient features in the local contour set of source identifiers to obtain the source identifier contour corresponding to the source identifier set on the purchased medicine.

[0028] Preferably, the information discrimination module includes:

[0029] The information segmentation unit is used to segment the source information into segments based on a preset lexicon to obtain the corresponding segment sequence, calculate the correlation between adjacent segments in the segment sequence, insert segmentation boundaries between adjacent segments with correlation less than a correlation threshold, and determine the corresponding information segment sequence based on all the segmentation boundaries contained in the segment sequence.

[0030] The stage determination unit is used to extract the corresponding information structure features in the information segment sequence, and match the information structure features with the standard information structure features corresponding to each circulation stage to determine the circulation stage corresponding to each information segment in the information segment sequence.

[0031] The level determination unit is used to determine the part of speech of each word segment contained in the information segment, extract the keyword segment corresponding to the keyword segment corresponding to the circulation stage based on the corresponding part of speech in the information segment, classify the keyword segment into categories, obtain the keyword segment set corresponding to each category, and determine the word segment level corresponding to each keyword segment contained in each category based on the word segment level determination standard corresponding to each category.

[0032] The graph generation unit is used to determine the word segment association relationship between all keyword segments based on the preset association relationship between different word segment categories and the word segment level corresponding to each keyword segment contained in each category, and to connect the corresponding keyword segments based on the word segment association relationship to obtain the associated word segment graph corresponding to the information segment;

[0033] The phrase matching unit is used to determine the association level corresponding to each keyword segment based on the associated phrase graph, sort the associated phrase graph corresponding to each information segment according to the information segment sequence to obtain the corresponding associated phrase graph sequence, and match the keyword segments with the same association level in the adjacent associated phrase graphs contained in the associated phrase graph sequence according to the association level to obtain the corresponding matched associated phrase group.

[0034] The coefficient calculation unit is used to calculate the content relevance between the matching related word segments contained in the matching related word segment group, determine the first relevance coefficient of the corresponding matching related word segment group based on the content relevance, calculate the second relevance coefficient of the corresponding adjacent related word segment graph based on the first relevance coefficient of all matching related word segment groups contained in the adjacent related word segment graph, and calculate the legality coefficient of the source information based on the second relevance coefficient of all adjacent related word segment graphs contained in the related word segment graph sequence.

[0035] The final judgment unit is used to determine that the source of the corresponding purchased drugs is legal when the legality coefficient is not less than the legality coefficient threshold; otherwise, it determines that the source of the corresponding purchased drugs is illegal.

[0036] Preferably, the inbound record terminal includes:

[0037] The number assignment module is used to determine the batch number and purchase time of the qualified drugs, and assign a unique number to each qualified drug according to the batch number and purchase time.

[0038] The inventory statistics module is used to count the personalized numbers of qualified medicines stored in the automatic medicine vending machine in real time and obtain the current inventory information of the automatic medicine vending machine.

[0039] The purchase record module is used to record the source information and purchase record of each qualified drug based on the personalized code corresponding to the qualified drug, and obtain the purchase record of each qualified drug.

[0040] Preferably, the end-to-end monitoring terminal includes:

[0041] The full-process monitoring module is used to monitor the qualified medicines throughout the entire process based on the high-definition camera installed inside the automatic medicine vending machine, and obtain the corresponding full-process monitoring video.

[0042] The video parsing module is used to parse the entire monitoring video to obtain the monitoring information corresponding to the qualified drugs.

[0043] Preferably, the video parsing module includes:

[0044] The video segmentation module is used to separate the full-process storage video and the full-process sales video corresponding to the qualified medicine from the full-process monitoring video.

[0045] The first parsing module is used to parse out the corresponding full-process storage information based on the full-process storage video and the storage environment management records of the automatic drug vending machine;

[0046] The second parsing module is used to parse out the corresponding full-process sales information based on the full-process sales video and the purchasing user information;

[0047] The information aggregation module is used to aggregate the entire storage information and the entire sales information to obtain the monitoring information corresponding to the qualified drugs.

[0048] Preferably, the first parsing module includes:

[0049] The video analysis unit is used to analyze the storage location time sequence of the corresponding qualified medicines based on the entire storage video.

[0050] The information correspondence unit is used to determine the storage environment data of the corresponding storage location at the corresponding time in the storage location time series based on the storage environment management record of the automatic drug vending machine, and generate the corresponding storage environment data sequence according to the storage location time series, and use the storage environment data sequence as the corresponding full-process storage information.

[0051] Preferably, the information updating terminal includes:

[0052] The inventory update module is used to update the current inventory information based on the sales information contained in the monitoring information to obtain the corresponding latest inventory information;

[0053] The circulation record module is used to generate corresponding drug circulation records based on the full-process storage information in the monitoring information, the sales information, and the purchase records.

[0054] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0055] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0056] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0057] Figure 1 This is a schematic diagram of a query and monitoring system for an automated drug vending machine according to an embodiment of the present invention;

[0058] Figure 2 This is a schematic diagram of an identification and traceability terminal in an embodiment of the present invention;

[0059] Figure 3 This is a schematic diagram of an identification module in an embodiment of the present invention;

[0060] Figure 4 This is a schematic diagram of a contour recognition unit in an embodiment of the present invention;

[0061] Figure 5 This is a schematic diagram of an information discrimination module in an embodiment of the present invention;

[0062] Figure 6 This is a schematic diagram of an inbound record terminal in an embodiment of the present invention;

[0063] Figure 7 This is a schematic diagram of a full-process monitoring terminal in an embodiment of the present invention;

[0064] Figure 8 This is a schematic diagram of a video parsing module in an embodiment of the present invention;

[0065] Figure 9 This is a schematic diagram of a first parsing module in an embodiment of the present invention;

[0066] Figure 10 This is a schematic diagram of an information update terminal in an embodiment of the present invention. Detailed Implementation

[0067] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0068] Example 1:

[0069] This invention provides a query and monitoring system for automated drug vending machines, with reference to... Figure 1 ,include:

[0070] The identification traceability terminal is used to authenticate the source of the purchased drugs based on the source identification of the purchased drugs;

[0071] The inventory entry record terminal is used to register the batch numbers of all qualified drugs that have passed source certification among all purchased drugs and to obtain the current inventory information of the drug vending machine and the purchase record of each qualified drug.

[0072] The end-to-end monitoring terminal is used to monitor the qualified drugs throughout the entire process and obtain corresponding monitoring information;

[0073] The information update terminal is used to update the current inventory information based on the monitoring information, and at the same time, generate corresponding drug circulation records based on the purchase records and the monitoring information.

[0074] In this embodiment, the source identifier is an identifier that contains information about the source of the corresponding purchased drugs.

[0075] In this embodiment, source authentication is to determine whether the source of the corresponding purchased drug is legal based on the source identifier of the corresponding purchased drug.

[0076] In this embodiment, batch number registration and warehousing means registering and warehousing the corresponding qualified drugs based on the batch number of the qualified drugs.

[0077] In this embodiment, qualified drugs are those purchased through source certification.

[0078] In this embodiment, the current inventory information is the current inventory information of the automatic drug vending machine, which includes information on the types and quantities of drugs.

[0079] In this embodiment, the purchase record refers to the purchase time and corresponding batch number of the qualified drugs, as well as other purchase-related information.

[0080] In this embodiment, the monitoring information is the information obtained after monitoring qualified drugs throughout the entire process.

[0081] In this embodiment, the drug circulation record is information generated based on purchase records and monitoring information that records the entire circulation process of drugs from production to sale to individual sellers.

[0082] The beneficial effects of the above technology are as follows: Based on the source identification of purchased medicines, the source of purchased medicines is certified and batch numbers are registered and stored in the warehouse. This allows for the acquisition of current inventory information of the automatic medicine vending machine and the purchase record of each qualified medicine, enabling the legality verification of medicine purchases and automatic information entry. Furthermore, by monitoring qualified medicines throughout the entire process, the storage management and monitoring of medicines are achieved. The current inventory information is updated in real time, and corresponding medicine circulation records are generated. This enables real-time and accurate management of the source of medicine purchases, sales, inventory, and expiration dates and batch numbers. Each automatic medicine vending machine can be monitored in real time, and detailed information on each sale is recorded, achieving transparent traceability of medicine circulation and enabling pharmacies to manage their medicine inventory.

[0083] Example 2:

[0084] Based on Example 1, the identification traceability terminal, refer to Figure 2 ,include:

[0085] The identification module is used to identify the source identification marks set on the purchased medicines and obtain the corresponding source information.

[0086] The information discrimination module is used to determine the legality of the source of the purchased drugs based on the source information;

[0087] The drug review module is used to determine that when the source of the purchased drug is determined to be legal, the corresponding purchased drug is a qualified drug that has passed source certification; otherwise, the corresponding purchased drug is a non-qualified drug that has not passed source certification.

[0088] The illegal alarm module is used to store the source information of the unqualified drugs when the purchased drugs are determined to be unqualified drugs that have not passed the source certification, and at the same time, issue a corresponding alarm signal.

[0089] In this embodiment, the source information is the information identified in the source identifier that represents the source process (production, transportation, distribution, etc.) of the corresponding purchased medicine.

[0090] In this embodiment, the substandard drug is the purchased drug that has not passed the source certification.

[0091] In this embodiment, the alarm signal is used to remind staff that the purchased medicines have not passed the source certification.

[0092] The beneficial effects of the above technologies are as follows: by using the source information in the source label set on the purchased drugs, the source of the purchased drugs can be certified, thereby verifying the legality of the source of the purchased drugs, ensuring the legality of the source of the purchased drugs, and realizing real-time and accurate management of the source of purchased drugs.

[0093] Example 3:

[0094] Based on Embodiment 2, the identification module refers to Figure 3 ,include:

[0095] The image acquisition unit is used to acquire an appearance image of the outer packaging of the purchased medicine at each preset angle to obtain a corresponding set of appearance images.

[0096] A contour recognition unit is used to identify the contour of the source identifier corresponding to the source identifier set on the purchased medicine in the set of appearance images;

[0097] An information identification unit is used to determine the corresponding source identification area in the set of appearance images based on the source identification outline, and to identify the source information of the purchased medicine in the source identification area.

[0098] In this embodiment, the preset angle is the pre-set shooting angle for capturing the outer packaging of the purchased medicines.

[0099] In this embodiment, the appearance image is an image of the outer packaging of the purchased medicine obtained from a preset angle.

[0100] In this embodiment, the set of appearance images is the set of appearance images.

[0101] In this embodiment, the source identifier outline is the outline of the source identifier set on the purchased medicines that is identified in the set of appearance images.

[0102] In this embodiment, the source identifier region is the image region corresponding to the source identifier determined in the set of appearance images based on the source identifier outline.

[0103] The beneficial effects of the above technology are as follows: Based on the appearance image of the outer packaging of the purchased medicine obtained from each preset angle, the outline corresponding to the source mark is identified, and the area corresponding to the source mark is further identified, thereby realizing the extraction of the source information contained in the source mark on the purchased medicine.

[0104] Example 4:

[0105] Based on Embodiment 3, the contour recognition unit, with reference to Figure 4 ,include:

[0106] The model building subunit is used to identify the appearance outline of the purchased medicine based on the appearance image, and to build a corresponding shape model of the purchased medicine based on the appearance outline.

[0107] The matrix determination subunit is used to determine the first point cloud data corresponding to the source identifier under the standard spatial shape, and to identify the corresponding second point cloud data based on the shape model of the purchased medicine, and to determine the corresponding first spatial shape transformation matrix based on the first point cloud data and the second point cloud data.

[0108] The point cloud transformation subunit is used to perform edge detection on the appearance image, identify the edge contour contained in the appearance image, determine the third point cloud data corresponding to the edge contour in the shape model of the purchased medicine based on the position of the edge contour in the corresponding appearance image, and determine the fourth point cloud data corresponding to the edge contour under the standard spatial shape based on the first spatial shape transformation matrix and the third point cloud data.

[0109] The feature extraction subunit is used to extract the shape features of the edge contour under the standard spatial shape based on the fourth point cloud data;

[0110] The contour sorting subunit is used to filter out the edge contours of the shape features that are consistent with the standard shape features corresponding to the source identifier in the appearance image, and use them as the corresponding source identifier local contours. According to the spatial relationship between the preset angles of the appearance images to which the source identifier local contours belong, the unit sorts and summarizes all the source identifier local contours in the appearance image set to obtain the corresponding source identifier local contour set.

[0111] A standard processing subunit is used to determine the light and shadow gradient features in the appearance image to which the local contour of the source identifier belongs, and at the same time, determine the visual effect evaluation value of the local contour of the source identifier. The light and shadow gradient features in the appearance image to which the local contour of the source identifier belongs corresponding to the maximum visual effect evaluation value are used as standard light and shadow gradient features. Based on the standard light and shadow gradient features, the remaining appearance images in the appearance image set other than the standard appearance images corresponding to the standard light and shadow gradient features are subjected to unified light and shadow feature processing to obtain the corresponding standard appearance image set.

[0112] The contour splicing subunit is used to determine the edges to be spliced ​​in all local contours of standard source identifiers contained in the standard appearance image set, determine the local splicing discrimination edges in the corresponding local contours of standard source identifiers based on the edges to be spliced, extract grayscale gradient features from the local splicing discrimination edges to obtain the corresponding local grayscale gradient features, and overlap and connect the local splicing discrimination edges with consistent local grayscale gradient features in the local contour set of source identifiers to obtain the source identifier contour corresponding to the source identifier set on the purchased medicine.

[0113] In this embodiment, the appearance outline refers to the outline of the shape of the purchased medicine.

[0114] In this embodiment, the shape model of the purchased medicines is a model that represents the three-dimensional shape of the purchased medicines based on their appearance contours.

[0115] In this embodiment, the standard spatial shape is a planar shape.

[0116] In this embodiment, the first point cloud data is the point cloud data of the source identifier in a standard spatial shape.

[0117] In this embodiment, the second point cloud data is the point cloud data representing the shape of the purchased drugs identified based on the shape model of the purchased drugs.

[0118] In this embodiment, the first spatial shape transformation matrix is ​​the transformation matrix corresponding to the transformation of the second point cloud data to the point cloud data corresponding to the standard spatial shape.

[0119] In this embodiment, the edge contour is the contour contained in the appearance image.

[0120] In this embodiment, the third point cloud data is the point cloud data corresponding to the shape model of the purchased medicine, which is determined based on the position of the edge contour in the appearance image.

[0121] In this embodiment, the fourth point cloud data is the point cloud data corresponding to the standard spatial shape after the third point cloud data is transformed based on the first spatial shape transformation matrix.

[0122] In this embodiment, the shape feature is the feature related to the contour shape corresponding to the edge contour extracted based on the fourth point cloud data under the standard spatial shape.

[0123] In this embodiment, the source identifier local contour set is a set of source identifier local contours obtained by sorting and summarizing all source identifier local contours contained in the appearance image set according to the spatial relationship between the preset angles corresponding to the appearance images to which the source identifier local contours belong.

[0124] In this embodiment, the local contour of the source identifier is the edge contour corresponding to the shape feature that is consistent with the standard shape feature corresponding to the source identifier and selected from the appearance image.

[0125] In this embodiment, the standard shape feature is the shape feature corresponding to the source identifier under the standard spatial shape.

[0126] In this embodiment, the standard appearance image set is a set of standard appearance images obtained by uniformly processing the remaining appearance images in the appearance image set, excluding the standard appearance images corresponding to the standard light and shadow gradient features, based on the standard light and shadow gradient features.

[0127] In this embodiment, the light and shadow gradient feature is the brightness gradient feature.

[0128] In this embodiment, determining the visual effect evaluation value of the local outline of the source identifier includes:

[0129]

[0130] In the formula, α is the visual effect evaluation value of the local outline of the source identifier, i is the i-th pixel point contained in the local outline of the source identifier, n is the total number of pixels contained in the local outline of the source identifier, and u i u represents the grayscale value of the i-th pixel contained in the local contour of the source identifier. max The maximum grayscale value is u. max The value is 255, o i o is the brightness value of the i-th pixel contained in the local contour of the source identifier. max This is the maximum brightness value, and o max The value is 100, p i p represents the chromaticity value of the i-th pixel contained in the local contour of the source identifier. max The maximum chromaticity value is p. max The value is 255;

[0131] For example, if the local outline of the source identifier contains 3 pixels, the grayscale value, brightness value, and chromaticity value of the first pixel are 50, 50, and 50 respectively, and the grayscale value, brightness value, and chromaticity value of the second pixel are 100, 100, and 100 respectively, then α is 0.45.

[0132] In this embodiment, the standard light and shadow gradient feature is the light and shadow gradient feature in the appearance image to which the local outline of the source identifier belongs, corresponding to the maximum visual effect evaluation value.

[0133] In this embodiment, the source identifier outline is the outline of the source identifier set on the purchased medicine after overlapping and connecting the local splicing discrimination edges with consistent local gray-scale gradient features in the source identifier local outline set.

[0134] In this embodiment, the edge to be stitched is the part of the local outline of the standard source identifier that overlaps with the edge of the appearance image.

[0135] In this embodiment, the local splicing discrimination edge is the edge of a preset length (specifically set according to the actual situation) that is adjacent to the edge to be spliced ​​in the local outline of the corresponding standard source identifier.

[0136] In this embodiment, the local grayscale gradient feature is the corresponding grayscale gradient feature obtained by extracting the grayscale gradient feature of the local splicing discrimination edge.

[0137] The beneficial effects of the above technology are as follows: A shape model of the purchased medicine is constructed based on its appearance image. Then, based on the spatial shape transformation matrix between the first point cloud data corresponding to the shape model and the second point cloud data corresponding to the source identifier in standard space, all edge contours identified on the purchased medicine are transformed into point cloud data corresponding to the standard spatial shape. This achieves the transformation of all contours identified on the purchased medicine into the standard spatial shape, improving the accuracy of source identifier contour recognition. Furthermore, based on the shape features corresponding to the point cloud data of the edge contours in standard space and the standard shape features corresponding to the source identifier, the corresponding local contour of the source identifier is identified among all edge contours. These local contours are then sorted, and based on the grayscale features between adjacent local contours, the overlapping parts between adjacent local contours are determined, thus achieving overlapping connection of the local contours and determining an accurate and complete source identifier contour. This provides an important foundation for the subsequent accurate identification of the source identifier region.

[0138] Example 5:

[0139] Based on Example 4, the information discrimination module refers to Figure 5 ,include:

[0140] The information segmentation unit is used to segment the source information into segments based on a preset lexicon to obtain the corresponding segment sequence, calculate the correlation between adjacent segments in the segment sequence, insert segmentation boundaries between adjacent segments with correlation less than a correlation threshold, and determine the corresponding information segment sequence based on all the segmentation boundaries contained in the segment sequence.

[0141] The stage determination unit is used to extract the corresponding information structure features in the information segment sequence, and match the information structure features with the standard information structure features corresponding to each circulation stage to determine the circulation stage corresponding to each information segment in the information segment sequence.

[0142] The level determination unit is used to determine the part of speech of each word segment contained in the information segment, extract the keyword segment corresponding to the keyword segment corresponding to the circulation stage based on the corresponding part of speech in the information segment, classify the keyword segment into categories, obtain the keyword segment set corresponding to each category, and determine the word segment level corresponding to each keyword segment contained in each category based on the word segment level determination standard corresponding to each category.

[0143] The graph generation unit is used to determine the word segment association relationship between all keyword segments based on the preset association relationship between different word segment categories and the word segment level corresponding to each keyword segment contained in each category, and to connect the corresponding keyword segments based on the word segment association relationship to obtain the associated word segment graph corresponding to the information segment;

[0144] The phrase matching unit is used to determine the association level corresponding to each keyword segment based on the associated phrase graph, sort the associated phrase graph corresponding to each information segment according to the information segment sequence to obtain the corresponding associated phrase graph sequence, and match the keyword segments with the same association level in the adjacent associated phrase graphs contained in the associated phrase graph sequence according to the association level to obtain the corresponding matched associated phrase group.

[0145] The coefficient calculation unit is used to calculate the content relevance between the matching related word segments contained in the matching related word segment group, determine the first relevance coefficient of the corresponding matching related word segment group based on the content relevance, calculate the second relevance coefficient of the corresponding adjacent related word segment graph based on the first relevance coefficient of all matching related word segment groups contained in the adjacent related word segment graph, and calculate the legality coefficient of the source information based on the second relevance coefficient of all adjacent related word segment graphs contained in the related word segment graph sequence.

[0146] The final judgment unit is used to determine that the source of the corresponding purchased drugs is legal when the legality coefficient is not less than the legality coefficient threshold; otherwise, it determines that the source of the corresponding purchased drugs is illegal.

[0147] In this embodiment, the information segment sequence is a sequence of information segments obtained by dividing the source information based on all the dividing boundaries contained in the word segment sequence, wherein each information segment consists of at least one word segment.

[0148] In this embodiment, the preset vocabulary is a vocabulary that is prepared in advance and contains all possible word segments in the source information.

[0149] In this embodiment, the word segment sequence is a sequence of word segments obtained by dividing the source information into segments based on a preset thesaurus.

[0150] In this embodiment, the correlation between adjacent word segments in the word segment sequence is calculated, including:

[0151] In the preset article library, determine the total number of first articles that only contain the first word segment (the first word segment among adjacent word segments), the total number of second articles that only contain the second word segment (the second word segment among adjacent word segments), and the total number of third articles that contain both the first word segment and the second word segment among adjacent word segments;

[0152] Based on the total number of the first, second, and third articles, as well as the total number of the fourth article in the preset article library, the correlation between adjacent word segments in the word segment sequence is calculated:

[0153]

[0154] In the formula, β is the relevance between adjacent word segments in the word segment sequence, N1 is the total number of the first article, N2 is the total number of the second article, N3 is the total number of the third article, and N4 is the total number of the fourth article;

[0155] For example, if N1 is 10, N2 is 10, N3 is 8, and N4 is 50, then β is 0.27.

[0156] In this embodiment, the relevance threshold is the maximum relevance required when a dividing line needs to be inserted between adjacent word segments.

[0157] In this embodiment, the circulation stages include, for example, production, transportation, and distribution.

[0158] In this embodiment, the information structure feature is the feature that represents the information structure in the information segment sequence. For example, "xxx Street, Chaoyang District, Beijing", the information structure feature corresponding to this information segment sequence is the address information down to the street level.

[0159] In this embodiment, the standard information structure feature is the standard information structure feature corresponding to the circulation stage.

[0160] In this embodiment, the segment level is determined based on the segment level determination standard corresponding to each category, which determines the level corresponding to each keyword segment contained in each category.

[0161] In this embodiment, the parts of speech include, for example, nouns and verbs.

[0162] In this embodiment, the keyword type is set according to the actual situation: for example, the keyword type corresponding to the production stage is noun.

[0163] In this embodiment, the keyword segment is the segment of words corresponding to the keywords of the corresponding circulation stage extracted from the information segment.

[0164] In this embodiment, the keyword segment set is the set of keyword segments corresponding to each category obtained after classifying the keyword segments. For example, the keyword segment set corresponding to address nouns (categories) and the keyword segment set corresponding to production raw material nouns (categories).

[0165] In this embodiment, the segment level determination standard is the segment level determination standard corresponding to different categories. For example, for address terms, the corresponding address terms are determined by the geographical range from large to small.

[0166] In this embodiment, the associated segment graph is a graph that represents the relationship between segments contained in the corresponding information segment after the corresponding keyword segments are associated and connected based on the segment association relationship.

[0167] In this embodiment, the preset association relationship is the pre-prepared association relationship between different word segment categories.

[0168] In this embodiment, the word segment association relationship is the association relationship between all keyword segments determined based on the preset association relationship between different word segment categories and the word segment level corresponding to each keyword segment contained in each category.

[0169] In this embodiment, the matching associated word segment group is a combination of matched associated word segments obtained by matching the keyword segments with the same association level in adjacent associated word segment maps contained in the associated word segment map sequence based on the association level.

[0170] In this embodiment, the association level is the number of the corresponding keyword segment in the association segment graph, determined from top to bottom.

[0171] In this embodiment, the associated word segment graph sequence is a sequence of associated word segment graphs obtained by sorting the associated word segment graphs corresponding to each information segment according to the information segment sequence.

[0172] In this embodiment, calculating the content relevance between the matching related word segments contained in the matching related word segment group includes:

[0173] The total number of articles in the preset article library is determined as follows: the fifth total number of articles that only contain the first segment of any two segments in the matching associated segment group (the first segment of any two segments in the matching associated segment group); the sixth total number of articles that only contain the second segment of any two segments in the adjacent segment group (the second segment of any two segments in the matching associated segment group); and the seventh total number of articles that simultaneously contain the first and second segments of the adjacent segment group.

[0174] Based on the total number of articles in the fifth, sixth, and seventh categories, as well as the total number of articles in the preset article library in the eighth category, the content relevance between the matching related word segments contained in the matching related word segment group is calculated:

[0175]

[0176] In the formula, δ represents the content relevance between the matching related word segments contained in the matching related word segment group, N5 represents the total number of fifth articles, N6 represents the total number of sixth articles, N7 represents the total number of seventh articles, and N8 represents the total number of eighth articles.

[0177] For example, if N5 is 10, N6 is 10, N7 is 8, and N8 is 50, then δ is 0.27.

[0178] In this embodiment, the relevance threshold is the maximum relevance required when a dividing line needs to be inserted between adjacent word segments.

[0179] In this embodiment, the first relevance coefficient of the corresponding matching related word segment group is determined based on the content relevance, that is: the average content relevance among all matching related word segments contained in the matching related word segment group is taken as the corresponding first relevance coefficient.

[0180] In this embodiment, the second relevance coefficient of the corresponding adjacent related word segment graph is calculated based on the first relevance coefficient corresponding to all matching related word segment groups contained in the adjacent related word segment graph, which is:

[0181] The average of the first relevance coefficients of all matching related word groups contained in the adjacent related word segment graph is used as the corresponding second relevance coefficient.

[0182] In this embodiment, the legitimacy coefficient of the source information is calculated based on the second relevance coefficients corresponding to all adjacent related word segment graphs contained in the related word segment graph sequence, which is:

[0183] The average of the second relevance coefficients of all adjacent related word segment graphs contained in the related word segment graph sequence is used as the legitimate coefficient of the source information.

[0184] In this embodiment, the legality coefficient threshold is the minimum legality coefficient corresponding to when the purchased drugs are determined to be from a legal source.

[0185] The beneficial effects of the above technology are as follows: After segmenting the source information based on a preset thesaurus, the division boundaries in the source information are determined based on the content relevance between adjacent segments. Based on the division boundaries, the source information is segmented to obtain corresponding information segment sequences. Then, based on the information structure features corresponding to the information segment sequences and the standard information structure features corresponding to each circulation stage, the circulation stage corresponding to each information segment is determined. Next, based on the keywords corresponding to each circulation stage, the information segments are segmented, and then categorized and their levels are determined. This provides important reference data for subsequently constructing the associated segment graph corresponding to the information segments. Then, based on the preset association relationships between keyword segments of different categories and the determined segment levels, the corresponding associated segment graph is constructed. This provides an important information foundation for subsequently judging the legality of source information based on the relevance between segments. Finally, based on the matching of segments of the same level in the adjacent associated segment graphs corresponding to adjacent information segments in the information segment sequence and the determination of content relevance, the legality of information in the source information is accurately judged, thereby also achieving accurate source certification of purchased drugs.

[0186] Example 6:

[0187] Based on Example 1, the warehousing record terminal, refer to Figure 6 ,include:

[0188] The number assignment module is used to determine the batch number and purchase time of the qualified drugs, and assign a unique number to each qualified drug according to the batch number and purchase time.

[0189] The inventory statistics module is used to count the personalized numbers of qualified medicines stored in the automatic medicine vending machine in real time and obtain the current inventory information of the automatic medicine vending machine.

[0190] The purchase record module is used to record the source information and purchase record of each qualified drug based on the personalized code corresponding to the qualified drug, and obtain the purchase record of each qualified drug.

[0191] In this embodiment, the batch number is the production batch number that comes with the product.

[0192] In this embodiment, the purchase time is the time when the qualified drug is determined to be a qualified drug.

[0193] In this embodiment, the personalized number is generated by sorting drugs with the same production batch number based on the purchase time, and each drug has a unique personalized number.

[0194] The beneficial effects of the above technology are as follows: based on the batch number and purchase time of each drug, a unique and personalized number is assigned to each purchased drug, which greatly facilitates the subsequent statistics of drug inventory and the recording of drug source information and purchase records.

[0195] Example 7:

[0196] Based on Example 1, the full-process monitoring terminal, refer to Figure 7 ,include:

[0197] The full-process monitoring module is used to monitor the qualified medicines throughout the entire process based on the high-definition camera installed inside the automatic medicine vending machine, and obtain the corresponding full-process monitoring video.

[0198] The video parsing module is used to parse the entire monitoring video to obtain the monitoring information corresponding to the qualified drugs.

[0199] In this embodiment, the full-process monitoring video is the video obtained by monitoring qualified medicines throughout the entire process using a high-definition camera installed inside the automatic medicine vending machine.

[0200] The beneficial effects of the above technologies are as follows: Based on high-definition cameras, qualified medicines can be monitored throughout the entire process, thereby realizing the monitoring of medicines from the time of purchase to the time of sale to individual sellers, providing an important data information foundation for realizing the transparency and traceability of the medicine circulation process.

[0201] Example 8:

[0202] Based on embodiment 7, the video parsing module, referencing Figure 8 ,include:

[0203] The video segmentation module is used to separate the full-process storage video and the full-process sales video corresponding to the qualified medicine from the full-process monitoring video.

[0204] The first parsing module is used to parse out the corresponding full-process storage information based on the full-process storage video and the storage environment management records of the automatic drug vending machine;

[0205] The second parsing module is used to parse out the corresponding full-process sales information based on the full-process sales video and the purchasing user information;

[0206] The information aggregation module is used to aggregate the entire storage information and the entire sales information to obtain the monitoring information corresponding to the qualified drugs.

[0207] In this embodiment, the full-process storage video refers to the monitoring video of the corresponding qualified drug storage process that is divided out from the full-process monitoring video.

[0208] In this embodiment, the full-process sales video refers to the monitoring video of the corresponding qualified drug sales process that is segmented from the full-process monitoring video.

[0209] In this embodiment, the storage environment management record is the environmental management data of different storage locations in the automatic medicine vending machine recorded according to time, such as the temperature control value and humidity control value within a certain time period.

[0210] In this embodiment, the full-process storage information refers to the information representing the storage environment management of the corresponding qualified drugs, which is parsed from the full-process storage video and the storage environment management records of the drug vending machine.

[0211] In this embodiment, the purchasing user information refers to the user information of the user who purchases the corresponding qualified medicine, such as payment account information.

[0212] In this embodiment, the full sales information refers to the information that ensures the sale of the corresponding qualified drugs, which is parsed from the full sales video and the information of the purchasing users.

[0213] The beneficial effects of the above technologies are as follows: by dividing the full-process monitoring video, corresponding full-process storage video and full-process sales video are obtained. Combined with the storage environment management records and purchase user information of the automatic drug vending machine, the classification, analysis and extraction of the drug storage process and sales process are realized. It also realizes the full-process tracking and monitoring of qualified drugs and the automatic entry and uploading of information, thereby strengthening the supervision of drugs in the automatic drug vending machine.

[0214] Example 9:

[0215] Based on Example 7, the first parsing module, referring to Figure 9 ,include:

[0216] The video analysis unit is used to analyze the storage location time sequence of the corresponding qualified medicines based on the entire storage video.

[0217] The information correspondence unit is used to determine the storage environment data of the corresponding storage location at the corresponding time in the storage location time series based on the storage environment management record of the automatic drug vending machine, and generate the corresponding storage environment data sequence according to the storage location time series, and use the storage environment data sequence as the corresponding full-process storage information.

[0218] In this embodiment, the storage location time series is the sequence parsed from the entire storage video, representing the corresponding storage location of the qualified drug in the automatic drug vending machine at different time periods.

[0219] In this embodiment, the storage environment data refers to the storage environment management data at the corresponding storage location at the corresponding time in the storage location time series determined based on the storage environment management records of the drug vending machine.

[0220] In this embodiment, the storage environment data sequence is a sequence of storage environment data obtained by sorting the storage environment data at the corresponding storage location at the corresponding time in the storage location time series based on the storage location time series.

[0221] The beneficial effects of the above technology are as follows: Based on the analysis of the storage video throughout the entire process, the storage location time series of the corresponding qualified drugs is obtained, which includes the storage location of the corresponding purchased drugs in different time periods. Then, combined with the storage environment data of the corresponding time period in the storage environment management record of the drug vending machine, the corresponding storage environment data sequence is generated, thereby obtaining the full storage information of the corresponding qualified drugs. This realizes the full tracking and monitoring of qualified drugs and the automatic entry and uploading of information, and strengthens the supervision of drugs in the drug vending machine.

[0222] Example 10:

[0223] Based on Example 1, the information update terminal refers to Figure 10 ,include:

[0224] The inventory update module is used to update the current inventory information based on the sales information contained in the monitoring information to obtain the corresponding latest inventory information;

[0225] The circulation record module is used to generate corresponding drug circulation records based on the full-process storage information in the monitoring information, the sales information, and the purchase records.

[0226] In this embodiment, the latest inventory information is the latest inventory information obtained after updating the current inventory information based on the sales information contained in the monitoring information.

[0227] The beneficial effects of the above technologies are as follows: the current inventory information is updated based on the sales information contained in the monitoring information, realizing the real-time automatic update of the inventory information; the corresponding qualified drug circulation records are generated based on the full storage information, sales information and purchase records contained in the monitoring information, realizing the full monitoring of the circulation of the corresponding drugs and strengthening the supervision of drugs in the automatic drug vending machine.

[0228] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A query supervision system of a medicine automatic vending machine, characterized by, The method comprises the following steps: An identification tracing end is used to identify the source of the purchased medicine based on the source identification of the purchased medicine; A storage record end is used to record the batch number of qualified medicine that has passed the source authentication among all the purchased medicine, obtain the current inventory information of the medicine vending machine and the purchase record of each qualified medicine; A whole-process monitoring end is used to monitor the whole process of the qualified medicine and obtain the corresponding monitoring information; An information updating end is used to update the current inventory information based on the monitoring information, and generate the corresponding medicine circulation record based on the purchase record and the monitoring information; The identification tracing end comprises: An identification recognition module is used to identify the source identification on the purchased medicine and obtain the corresponding source information; An information discrimination module is used to discriminate the source legality of the purchased medicine based on the source information; A medicine review module is used to determine that the corresponding purchased medicine is a qualified medicine that has passed the source authentication when the purchased medicine is determined to be legal in source, otherwise, it is determined that the corresponding purchased medicine is unqualified medicine that has not passed the source authentication; An illegal alarm module is used to store the source information of the unqualified medicine when there is purchased medicine that is determined to be unqualified medicine that has not passed the source authentication, and issue a corresponding alarm signal; The identification recognition module comprises: An image acquisition unit is used to acquire the appearance image of the outer package of the purchased medicine at each preset angle to obtain a corresponding appearance image set; A contour recognition unit is used to identify the source identification contour of the source identification provided on the purchased medicine in the appearance image set; An information recognition unit is used to determine the source identification area in the appearance image set based on the source identification contour, and identify the source information of the purchased medicine in the source identification area; The contour recognition unit comprises: A model construction subunit is used to identify the appearance contour of the purchased medicine based on the appearance image, and construct a corresponding purchased medicine shape model based on the appearance contour; A matrix determination subunit is used to determine the first point cloud data corresponding to the source identification in the standard space shape, and identify the second point cloud data corresponding to the purchased medicine shape model based on the first point cloud data and the second point cloud data, and determine the first space shape transformation matrix based on the first point cloud data and the second point cloud data; A point cloud transformation subunit is used to perform edge detection on the appearance image, identify the edge contour contained in the appearance image, determine the third point cloud data of the edge contour in the purchased medicine shape model based on the position of the edge contour in the corresponding appearance image, and determine the fourth point cloud data of the edge contour in the standard space shape based on the first space shape transformation matrix and the third point cloud data; A feature extraction subunit is used to extract the shape features of the edge contour in the standard space shape based on the fourth point cloud data. The contour arrangement subunit is configured to screen out edge contours corresponding to shape features consistent with standard shape features corresponding to the source mark in the appearance images as corresponding source mark local contours, sort and aggregate all source mark local contours contained in the appearance image set according to the spatial relationship between preset angles corresponding to the appearance images to which the source mark local contours belong, and obtain a corresponding source mark local contour set; The standard processing subunit is configured to determine a light and shadow transition feature in the appearance image to which the source mark local contour belongs, determine a visual effect evaluation value of the source mark local contour, take a light and shadow transition feature in an appearance image to which a source mark local contour corresponding to a maximum visual effect evaluation value belongs as a standard light and shadow transition feature, perform light and shadow feature uniform processing on appearance images in the appearance image set except for a standard appearance image corresponding to the standard light and shadow transition feature based on the standard light and shadow transition feature, and obtain a corresponding standard appearance image set; The contour splicing subunit is configured to determine a to-be-spliced edge in all standard source mark local contours contained in the standard appearance image set, determine a local splicing judgment edge in a corresponding standard source mark local contour based on the to-be-spliced edge, extract a local gray level transition feature of the local splicing judgment edge, overlap and connect local splicing judgment edges with the same local gray level transition feature in the source mark local contour set, and obtain a source mark contour corresponding to the source mark arranged on the purchased medicine. The standard spatial shape is a planar shape. The first spatial shape transformation matrix is a transformation matrix corresponding to the conversion of the second point cloud data to the point cloud data corresponding to the standard spatial shape. The light and shadow transition feature is a brightness transition feature.

2. The query monitoring system of the automatic medicine vending machine according to claim 1, wherein, The information discrimination module includes: The information division unit is configured to divide the source information into word segments based on a preset word library, obtain a corresponding word segment sequence, calculate the correlation between adjacent word segments in the word segment sequence, insert a division boundary between adjacent word segments with a correlation less than a correlation threshold, and determine a corresponding information segment sequence based on all division boundaries contained in the word segment sequence. The stage determination unit is configured to extract information structure features corresponding to the information segment sequence, match the information structure features with standard information structure features corresponding to each circulation stage, and determine a circulation stage corresponding to each information segment contained in the information segment sequence. The level determination unit is configured to determine the part of speech of each word segment contained in the information segment, extract a key word segment corresponding to a key part of speech corresponding to the corresponding circulation stage in the information segment based on the corresponding part of speech, classify the key word segment, obtain a key word segment set corresponding to each category, and determine a word segment level corresponding to each key word segment contained in each category based on a word segment level determination standard corresponding to each category. The atlas generating unit is configured to determine word segment association relationships among all the keyword segments based on preset association relationships among different word segment categories and word segment levels corresponding to each keyword segment included in each category, and to obtain an associated word segment atlas corresponding to the information segment by associating and connecting the corresponding keyword segments based on the word segment association relationships; The word group matching unit is configured to determine an association level corresponding to each keyword segment based on the associated word segment atlas, to sort the associated word segment atlas corresponding to each information segment according to the information segment sequence to obtain a corresponding associated word segment atlas sequence, and to perform corresponding matching on keyword segments with the same association level in adjacent associated word segment atlases included in the associated word segment atlas sequence based on the association level to obtain a corresponding matching associated word segment group. The coefficient calculating unit is configured to calculate content correlation degrees between matching associated word segments included in the matching associated word segment group, to determine a first correlation degree coefficient of the corresponding matching associated word segment group based on the content correlation degrees, to calculate a second correlation degree coefficient of the corresponding adjacent associated word segment atlas based on first correlation degree coefficients corresponding to all the matching associated word segment groups included in the adjacent associated word segment atlas, and to calculate a legitimacy coefficient of the source information based on second correlation degree coefficients corresponding to all the adjacent associated word segment atlases included in the associated word segment atlas sequence. The final judgment unit is configured to determine that the source of the purchased medicine is legitimate when the legitimacy coefficient is not less than a legitimacy coefficient threshold, and to determine that the source of the purchased medicine is not legitimate otherwise.

3. The query monitoring system of claim 1, wherein, The storage record end includes: The number assignment module is configured to determine a factory batch number and a purchase time of the qualified medicine, and to assign a personalized number corresponding to each qualified medicine according to the factory batch number and the purchase time of the qualified medicine. The inventory statistical module is configured to statistically obtain the personalized numbers of the qualified medicine stored in the medicine vending machine in real time to obtain current inventory information of the medicine vending machine. The purchase record module is configured to record source information and purchase records of each qualified medicine based on the personalized codes corresponding to the qualified medicine to obtain the purchase records of each qualified medicine.

4. The query monitoring system of claim 1, wherein, The whole-process monitoring end includes: The whole-process monitoring module is configured to perform whole-process monitoring on the qualified medicine based on a high-definition camera arranged inside the medicine vending machine to obtain a corresponding whole-process monitoring video. The video analysis module is configured to analyze the whole-process monitoring video to obtain monitoring information corresponding to the qualified medicine.

5. The query monitoring system of claim 4, wherein, The video analysis module includes: The video division module is configured to divide the whole-process monitoring video into a whole-process storage video and a whole-process sales video corresponding to the qualified medicine. The first analysis module is configured to analyze corresponding whole-process storage information based on the whole-process storage video and storage environment management records of the medicine vending machine. The second analysis module is configured to analyze corresponding whole-process sales information based on the whole-process sales video and purchase user information. The information summarizing module is configured to summarize the whole-process storage information and the whole-process sales information to obtain the monitoring information corresponding to the qualified medicine.

6. The query monitoring system of claim 5, wherein, The first analysis module includes: a video analysis unit, configured to analyze the whole-storage video to obtain a storage location time sequence of the qualified medicine; an information corresponding unit, configured to determine storage environment data of a storage location at a corresponding time in the storage location time sequence based on storage environment management records of the medicine vending machine, and generate a corresponding storage environment data sequence according to the storage location time sequence, and take the storage environment data sequence as corresponding whole-storage information.

7. The query monitoring system of claim 1, wherein, The information updating end comprises: an inventory updating module, configured to update the current inventory information based on the sales information contained in the monitoring information, and obtain corresponding latest inventory information; a circulation record module, configured to generate a corresponding medicine circulation record based on the whole-storage information, the sales information and the purchase record in the monitoring information.

Images