Commodity management method and device, storage medium and computer equipment

Through automated product management methods, image recognition and feature extraction models are used to solve the problems of out-of-stock and difficult to find timely arrangement of goods in the existing technology, accurate statistics and analysis of product inventory information are achieved, and retail management efficiency is improved. and accuracy.

CN118917777APending Publication Date: 2024-11-08DONGGUAN ZKTECO ELECTRONICS TECH
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411176492.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-26
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

In the prior art, problems that are difficult to detect and solve in a timely manner when out of stock or when goods are placed in chaos, lead to lagging information feedback and affecting retail management efficiency and accuracy.

Method used

Through a product management method, in response to product management requests, the product detection strategy and detection area are determined, and the product type and booth information are counted using image recognition and feature extraction models to match product inventory information, including out-of-stock, wrong placement and in-stock information. , realize automated product detection and matching.

Benefits of technology

Accurately count and analyze commodity inventory information, reduce human errors and negligence, ensure the accuracy and reliability of information, and reduce the time and labor costs of manual patrol inspections.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917777A_ABST
    Figure CN118917777A_ABST
Patent Text Reader

Abstract

According to the commodity management method and device, the storage medium and the computer equipment provided by the invention, in response to the commodity management request, the commodity detection strategy of the commodity management request and the corresponding commodity detection area are determined; if the commodity detection strategy comprises a commodity type matching mode, obtaining commodity type configuration information of a commodity detection area, and carrying out statistics on commodity type information in the commodity detection area to obtain a commodity type statistical result; matching the commodity type statistical result with the commodity type configuration information to obtain commodity inventory information of the commodity detection area, the commodity inventory information including commodity wrong placement information and stockout information; if the commodity detection strategy comprises a commodity booth matching mode, obtaining commodity booth configuration information of the commodity detection area; and determining commodity booth information in the commodity detection area, and matching the commodity booth information with the commodity booth configuration information to obtain commodity inventory information of the commodity detection area. Therefore, the time and the labor cost of manual patrol inspection are greatly reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of smart retail technology, and in particular to a commodity management method, device, storage medium and computer equipment. Background Art

[0002] With the development of technology and the change of consumer expectations, smart retail has emerged and become an important development trend in modern retail. The current retail sales model is mainly divided into two types: open space and closed space. In the open space, customers can freely take and put goods. Although this model provides greater shopping convenience, it also leads to randomness and instability in the placement of goods. Customers' random operations cause the placement of goods to change frequently, which brings great management challenges to retailers. In contrast, the closed space sales model is controlled by the salesperson to take and put, and the placement of goods remains relatively stable, which is relatively simple to manage.

[0003] However, whether in open or enclosed spaces, the traditional merchandise management model has significant shortcomings. Supermarkets usually rely on sales staff to conduct manual patrol inspections to ensure that the merchandise is neatly displayed and the inventory is sufficient. This is not only a time-consuming task, but also has the problem of delayed information feedback, making it difficult to detect and solve out-of-stock or chaotic merchandise placement in a timely manner. Summary of the invention

[0004] The purpose of this application is to solve at least one of the above-mentioned technical defects, especially the technical defects in the prior art that it is difficult to promptly discover and solve the problem when the goods are out of stock or the goods are placed in a disorderly manner.

[0005] In a first aspect, the present application provides a commodity management method, the method comprising:

[0006] In response to the commodity management request, determining a commodity detection strategy and a corresponding commodity detection area of ​​the commodity management request;

[0007] If the commodity detection strategy includes a commodity category matching method, then obtain commodity category configuration information of the commodity detection area, and count commodity category information in the commodity detection area to obtain commodity category statistics results;

[0008] Match the commodity category statistics with the commodity category configuration information to obtain commodity inventory information in the commodity detection area, the commodity inventory information includes commodity misplacement information and out-of-stock information;

[0009] If the product detection strategy includes a product booth matching method, then obtain the product booth configuration information of the product detection area;

[0010] The commodity booth information in the commodity detection area is determined and matched with the commodity booth configuration information to obtain the commodity inventory information of the commodity detection area.

[0011] In one embodiment, the step of collecting information about commodity categories in the commodity detection area to obtain commodity category statistics includes:

[0012] Cropping the image of the commodity detection area to obtain multiple commodity images in the commodity detection area;

[0013] Input each product image into a predetermined feature extraction model to obtain product features of each product image;

[0014] According to the characteristics of each commodity, the commodity category corresponding to each commodity in the commodity detection area is determined to obtain commodity category statistics results.

[0015] In one embodiment, if the commodity category statistics result only includes commodity categories in the commodity detection area, the step of matching the commodity category statistics result with the commodity category configuration information to obtain commodity inventory information in the commodity detection area includes:

[0016] Match the commodity category statistics results with the commodity category configuration information;

[0017] If there is a commodity type in the commodity type statistics result that is different from the commodity type configuration information, then the commodity misplacement information is determined based on the commodity type;

[0018] If the commodity category statistics result lacks the commodity category in the commodity category configuration information, out-of-stock information is determined based on the commodity category to obtain commodity inventory information.

[0019] In one of the embodiments, the product inventory information also includes insufficient inventory information;

[0020] If the commodity category statistics result includes the commodity categories and the corresponding commodity quantities in the commodity detection area, the steps of matching the commodity category statistics result with the commodity category configuration information to obtain out-of-stock information and commodity misplacement information of the commodity detection area include:

[0021] Match the commodity category statistics results with the commodity category configuration information;

[0022] If there is a commodity category whose commodity quantity is not zero in the commodity category configuration information, and the commodity quantity of the commodity category in the commodity category statistics result is zero, out-of-stock information is determined based on the commodity category;

[0023] If the commodity category statistics result includes commodity categories that are not listed in the commodity category configuration information, then the commodity misplacement information is determined based on the commodity category;

[0024] If for each commodity category in the commodity category statistics result, the commodity quantity of the commodity category is not zero and is less than the commodity quantity of the commodity category in the commodity category configuration information, then the insufficient inventory information is determined according to the commodity category to obtain the commodity inventory information.

[0025] In one embodiment, the commodity booth information includes actual commodity types and actual booth locations in the commodity detection area, and the commodity booth configuration information includes preset commodity types and preset booth locations in the commodity detection area;

[0026] The steps of determining the commodity booth information in the commodity detection area and matching it with the commodity booth configuration information to obtain the commodity inventory information in the commodity detection area include:

[0027] For each preset booth position in the product booth configuration information, if the preset booth position does not have a corresponding actual booth position in the product detection area, out-of-stock information is determined based on the preset booth position and its corresponding preset product type;

[0028] For each preset booth position in the merchandise booth configuration information, if there is a corresponding actual booth position in the merchandise detection area for the preset booth position, and the preset merchandise type corresponding to the preset booth position is different from the actual merchandise type corresponding to the actual booth position, then the merchandise misplacement information is determined based on the preset booth position and its corresponding preset merchandise type to obtain the merchandise inventory information.

[0029] In one embodiment, the process of determining that the preset commodity type corresponding to the preset booth position is different from the actual commodity type corresponding to the actual booth position includes:

[0030] Determine a preset product image corresponding to the preset booth position, and determine an actual product image corresponding to the actual booth position;

[0031] An IOU value between the preset product image and the actual product image is calculated. When the IOU value is less than a preset threshold, the preset product type corresponding to the preset booth position is different from the actual product type corresponding to the actual booth position.

[0032] In one embodiment, the commodity booth information includes actual commodity types and their actual booth order in the commodity detection area, and the commodity booth configuration information includes preset commodity types and their preset booth order in the commodity detection area;

[0033] The steps of determining the commodity booth information in the commodity detection area and matching it with the commodity booth configuration information to obtain the commodity inventory information in the commodity detection area include:

[0034] Determine a first sequence corresponding to the commodity booth information according to actual commodity types and actual booth sequence in the commodity detection area;

[0035] Determine a second sequence corresponding to the commodity booth configuration information according to the preset commodity types and the preset booth sequence in the commodity detection area;

[0036] Find the longest common subsequence between the first sequence and the second sequence;

[0037] Compare the longest common subsequence with the second sequence, obtain the commodity types that exist in the second sequence but not in the longest common subsequence, and determine out-of-stock information;

[0038] The longest common subsequence is compared with the first sequence to obtain the commodity types that exist in the first sequence but not in the longest common subsequence, and the commodity misplacement information is determined to obtain the commodity inventory information.

[0039] In a second aspect, the present application provides a commodity management device, the device comprising:

[0040] A commodity detection strategy determination module, used to determine the commodity detection strategy and corresponding commodity detection area of ​​the commodity management request in response to the commodity management request;

[0041] A commodity category statistical result determination module is used to obtain commodity category configuration information of the commodity detection area if the commodity detection strategy includes a commodity category matching method, and to count commodity category information in the commodity detection area to obtain commodity category statistical results;

[0042] A first commodity inventory information determination module is used to match the commodity category statistics result with the commodity category configuration information to obtain the commodity inventory information of the commodity detection area, where the commodity inventory information includes commodity misplacement information and out-of-stock information;

[0043] A commodity booth configuration information acquisition module, used to acquire commodity booth configuration information of a commodity detection area if the commodity detection strategy includes a commodity booth matching method;

[0044] The second commodity inventory information determination module is used to determine the commodity booth information in the commodity detection area and match it with the commodity booth configuration information to obtain the commodity inventory information of the commodity detection area.

[0045] In a third aspect, the present application provides a storage medium: the storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the commodity management method in any of the above embodiments.

[0046] In a fourth aspect, the present application provides a computer device, comprising: one or more processors, and a memory;

[0047] The memory stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the steps of the commodity management method in any of the above embodiments are executed.

[0048] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0049] In the commodity management method, device, storage medium and computer equipment provided in the present application, in response to a commodity management request, the commodity detection strategy and the corresponding commodity detection area of ​​the commodity management request are determined; if the commodity detection strategy includes a commodity type matching method, the commodity type configuration information of the commodity detection area is obtained, and the commodity type information in the commodity detection area is counted to obtain the commodity type statistical results; the commodity type statistical results are matched with the commodity type configuration information to obtain the commodity inventory information of the commodity detection area, and the commodity inventory information includes the commodity misplacement information and out-of-stock information; if the commodity detection strategy includes a commodity booth matching method, the commodity booth configuration information of the commodity detection area is obtained; the commodity booth information in the commodity detection area is determined, and matched with the commodity booth configuration information to obtain the commodity inventory information of the commodity detection area. Based on the method of commodity type matching and booth matching, the commodity inventory information can be accurately counted and analyzed to reduce human errors and negligence. At the same time, the method uses digital means to automatically compare commodity types and booth configurations to ensure the accuracy and reliability of information. In this way, through the automated commodity detection and matching process, the time and manpower cost of manual patrol inspections are greatly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0051] Figure 1 A flowchart of a commodity management method provided in an embodiment of the present application;

[0052] Figure 2 A schematic diagram of the structure of a commodity management device provided in an embodiment of the present application;

[0053] Figure 3 A schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0055] The present application provides a commodity management method. The following embodiments are described by taking the method applied to a computer device as an example. It can be understood that the computer device can be any device with data processing functions, including but not limited to a single server, a server cluster, a personal laptop computer, a desktop computer, etc. Figure 1 As shown, the method includes:

[0056] S101: In response to a commodity management request, determine a commodity detection strategy and a corresponding commodity detection area of ​​the commodity management request.

[0057] Among them, the commodity management request is used to perform certain management operations on the commodity, such as checking the inventory of the commodity, detecting the location and type of the commodity, and the information of the shelf. The commodity detection strategy is a specified set of rules or methods for identifying and verifying commodity information. Different commodities may have different detection strategies, which can be specified by merchants, administrators or automated tasks by initiating commodity management requests. For example, the commodity detection area and the corresponding commodity detection strategy can be set, and the commodity management request can be initiated based on this. Therefore, by parsing the commodity management request, the commodity detection strategy and the corresponding commodity detection area carried by the request can be determined.

[0058] A commodity detection area refers to one or more specific areas delineated based on the actual commodity placement and predefined matching strategies. These areas are used to accurately locate and detect information such as the status, inventory, and placement of commodities. For example, in a retail store, warehouse, or any other sales venue, a commodity detection area can be the specific location where the commodity is actually displayed or stored, including shelves, cabinets, display stands, etc. According to the matching strategy and the actual commodity placement, the store or warehouse is divided into one or more specific detection areas. These areas can be the entire shelf, several layers on the shelf, or even a specific area range. There are no specific restrictions on the shape and size of the area, but they must meet the requirements of the matching strategy to ensure the effectiveness and accuracy of the detection. Each delineated detection area corresponds to a specific matching strategy. This means that different areas may adopt different detection methods and standards. For example, a commodity area with a high inventory turnover rate may require more frequent inventory detection and replenishment reminders, while a commodity area with a low inventory turnover rate may pay more attention to the neatness and compliance of the commodity placement.

[0059] In this step, first receive a product management request from the user, which contains the type of product management operation the user wants to perform and any specific requirements or parameters that may be attached. Parse the received product management request to extract key information, including the product detection strategy specified by the user, the range of the detection area, the detection time interval, and any other parameters or conditions related to the request. Determine the product detection strategy to be used in the current request, which will guide the subsequent product detection process, including how to identify the product, verify product information, and how to process the detected data. According to the specified product detection area range, these areas are determined as the target areas to be detected by the current request. These areas should match the matching strategy to ensure the effectiveness and accuracy of the detection.

[0060] Further, when determining the commodity detection strategy and commodity detection area, it is possible to verify whether the specified commodity detection strategy and commodity detection area are valid and reasonable. For example, whether the commodity detection strategy exists in the predefined strategy set, whether the commodity detection area covers the valid commodity placement position, and whether there is a logical conflict between the commodity detection strategy and the commodity detection area. If any problems or inconsistencies are found, it may be necessary to return an error message or request a correction. Once the validity of the specified commodity detection strategy is verified, it is determined as the strategy to be used by the current request.

[0061] S102: If the commodity detection strategy includes a commodity category matching method, commodity category configuration information of the commodity detection area is obtained, and commodity category information in the commodity detection area is counted to obtain commodity category statistics results.

[0062] Among them, the commodity category matching method is used to match the detected commodity category with the predefined commodity category, which can be based on the commodity barcode, RFID tag, image recognition, or any other technology that can uniquely or roughly determine the commodity category. The commodity category configuration information is information about the distribution and expected configuration of commodity categories in the commodity detection area, which may include which types of commodities should appear in the area, the quantity limit of each commodity, or the placement requirements of specific commodity categories. The commodity category statistical result is the result obtained by counting the actual commodity category information in the commodity detection area, which may include the distribution of various commodity categories in the area.

[0063] In this step, first, when the commodity detection strategy is parsed and retrieved to determine the commodity category matching method, it is necessary to access the database or configuration file storing the commodity category configuration information. The commodity category configuration information corresponding to the area can be retrieved according to the identifier of the commodity detection area, such as the shelf number, area code, etc. Various detection technologies, such as RFID scanning, image recognition, barcode reading, etc., are used to detect the commodities in the commodity detection area. The detected commodities are identified and matched with the predefined commodity categories. The quantity of each commodity is counted, and the distribution of commodity categories is recorded, so as to obtain the commodity category statistics.

[0064] S103: Match the commodity category statistics result with the commodity category configuration information to obtain commodity inventory information of the commodity detection area, where the commodity inventory information includes commodity misplacement information and out-of-stock information.

[0065] Among them, the commodity inventory information is information comprehensively derived based on the commodity category statistics and commodity category configuration information, which reflects the inventory status of commodities in the commodity detection area. Specifically, the inventory information especially includes commodity misplacement information and out-of-stock information. Commodity misplacement information refers to commodities that do not conform to the commodity category configuration information in the commodity detection area. That is, some commodities are mistakenly placed in areas where they should not appear, or commodity categories that should not exist appear in certain areas. Out-of-stock information refers to the complete lack of certain commodity categories that should exist in the commodity detection area.

[0066] In this step, the commodity category statistics are matched with the commodity category configuration information to check the differences between the commodity category statistics and the commodity category configuration information. Specifically, each commodity category defined in the commodity category configuration information is traversed. For each commodity category, the corresponding entry is searched in the commodity category statistics, and the corresponding out-of-stock information and commodity misplacement information are generated according to the search result.

[0067] S104: If the commodity detection strategy includes a commodity booth matching method, obtain commodity booth configuration information of the commodity detection area.

[0068] Among them, the product booth matching method is a process of matching the actual detected product booth with the predefined product booth configuration, aiming to ensure that the display of the products in the sales area meets the established standards. The product booth configuration information is the specific configuration information of each booth in the product detection area, which may include detailed information such as the type, quantity, and placement of the products to be displayed on each booth.

[0069] In this step, first, it is necessary to confirm whether the current product detection strategy includes a product booth matching method, which can be determined by checking a specific field, flag bit or conditional expression in the strategy data. If it is included, the system or database storing the product booth configuration information is accessed, and the corresponding product booth configuration information is retrieved from the configuration information source according to the identifier of the product detection area, such as the area code, shelf number, etc.

[0070] S105: Determine the commodity booth information in the commodity detection area, and match it with the commodity booth configuration information to obtain commodity inventory information in the commodity detection area.

[0071] Among them, product booth information refers to the actual status information of products at each booth in the product inspection area, including product type, quantity, placement, etc., which can be obtained through remote monitoring and other means.

[0072] In this step, visual acquisition equipment, such as cameras, scanners, etc., can be used to shoot or scan the commodity booths in the detection area. Through computer vision technologies such as image processing and object recognition, key information such as the type, quantity, and placement of the goods on the booths can be extracted from the shot or scanned images. The collected commodity booth information is compared with the predefined commodity booth configuration information one by one, and the comparison content includes whether the type, quantity, and placement of the goods on the booth are consistent. The differences between the booth information and the configuration information are identified, and these differences are important components of the commodity inventory information. Based on the matching results, the inventory information of the commodity detection area is generated. The inventory information should include out-of-stock information, that is, which goods are missing, and wrong product placement information, that is, which goods are placed in the wrong position or type.

[0073] In the above embodiment, in response to a commodity management request, the commodity detection strategy and the corresponding commodity detection area of ​​the commodity management request are determined; if the commodity detection strategy includes a commodity type matching method, the commodity type configuration information of the commodity detection area is obtained, and the commodity type information in the commodity detection area is counted to obtain the commodity type statistical results; the commodity type statistical results are matched with the commodity type configuration information to obtain the commodity inventory information of the commodity detection area, and the commodity inventory information includes commodity misplacement information and out-of-stock information; if the commodity detection strategy includes a commodity booth matching method, the commodity booth configuration information of the commodity detection area is obtained; the commodity booth information in the commodity detection area is determined, and matched with the commodity booth configuration information to obtain the commodity inventory information of the commodity detection area. Based on the commodity type matching and booth matching methods, the commodity inventory information can be accurately counted and analyzed to reduce human errors and negligence. At the same time, the method uses digital means to automatically compare commodity types and booth configurations to ensure the accuracy and reliability of information. In this way, through the automated commodity detection and matching process, the time and manpower cost of manual patrol inspections are greatly reduced.

[0074] In one embodiment, the step of collecting information about commodity categories in the commodity detection area to obtain commodity category statistics includes:

[0075] Cropping the image of the commodity detection area to obtain multiple commodity images in the commodity detection area;

[0076] Input each product image into a predetermined feature extraction model to obtain product features of each product image;

[0077] According to the characteristics of each commodity, the commodity category corresponding to each commodity in the commodity detection area is determined to obtain commodity category statistics results.

[0078] Image cropping refers to the process of separating the area of ​​interest from the original image to remove background noise and unnecessary parts, and only retain the product image that needs to be analyzed. Product image refers to the image of a single product extracted from the product detection area after cropping. The feature extraction model is a pre-trained machine learning or deep learning model that is used to extract key features from the input product image. These features can describe the shape, color, texture and other attributes of the product, which helps to identify and classify. Product features refer to numerical values ​​or vectors extracted from product images by feature extraction models that can represent product attributes.

[0079] Specifically, the original product detection area image is cropped using image processing technology to separate the image of each product, and ensure that the cropped product image is clear and complete, without background noise or other irrelevant information. Each cropped product image is input into a predetermined feature extraction model, which automatically extracts key features from the image. These features are expressed in the form of numerical values ​​or vectors and can describe the unique attributes of the product. Then, a classification algorithm or classifier can be used to extract the product features, and the category corresponding to each product can be determined based on the product features, thereby obtaining the product category statistics.

[0080] In this embodiment, the feature extraction model can more accurately identify and classify commodity types, reducing misjudgment and missed judgment. The automated image cropping, feature extraction and classification process can significantly improve the efficiency of commodity detection and reduce manual intervention and time consumption.

[0081] In one embodiment, if the commodity category statistics result only includes commodity categories in the commodity detection area, the step of matching the commodity category statistics result with the commodity category configuration information to obtain commodity inventory information in the commodity detection area includes:

[0082] Match the commodity category statistics results with the commodity category configuration information;

[0083] If there is a commodity type in the commodity type statistics result that is different from the commodity type configuration information, then the commodity misplacement information is determined based on the commodity type;

[0084] If the commodity category statistics result lacks the commodity category in the commodity category configuration information, out-of-stock information is determined based on the commodity category to obtain commodity inventory information.

[0085] Specifically, first, the statistical results of commodity categories are compared with the configuration information of commodity categories one by one to find out the differences between the two. During the comparison process, if it is found that there are commodity categories in the statistical results that are not in the configuration information, these commodities are recorded as misplaced commodities and corresponding misplaced information is generated. If some commodity categories listed in the configuration information are not found in the statistical results, these commodities are marked as out of stock and out of stock information is generated.

[0086] In an example, if the product category configuration information includes product categories a, b, and c, and the product category statistics result includes product categories b, c, and d, it means that product category a is out of stock and product category d is misplaced.

[0087] In this embodiment, by automatically matching the statistical results of commodity categories with the configuration information, the time and error rate of manual verification can be greatly reduced, thereby improving the accuracy and efficiency of commodity management.

[0088] In one embodiment, the commodity inventory information further includes insufficient inventory information;

[0089] If the commodity category statistics result includes the commodity categories and the corresponding commodity quantities in the commodity detection area, the steps of matching the commodity category statistics result with the commodity category configuration information to obtain out-of-stock information and commodity misplacement information of the commodity detection area include:

[0090] Match the commodity category statistics results with the commodity category configuration information;

[0091] If there is a commodity category whose commodity quantity is not zero in the commodity category configuration information, and the commodity quantity of the commodity category in the commodity category statistics result is zero, out-of-stock information is determined based on the commodity category;

[0092] If the commodity category statistics result includes commodity categories that are not listed in the commodity category configuration information, then the commodity misplacement information is determined based on the commodity category;

[0093] If for each commodity category in the commodity category statistics result, the commodity quantity of the commodity category is not zero and is less than the commodity quantity of the commodity category in the commodity category configuration information, then the insufficient inventory information is determined according to the commodity category to obtain the commodity inventory information.

[0094] Among them, insufficient inventory information refers to the information generated when the existing quantity of a certain commodity in the commodity detection area is lower than its preset minimum inventory quantity or the quantity specified in the configuration information, indicating that the inventory level of the commodity is insufficient to meet current or expected demand.

[0095] Specifically, first, a comprehensive comparison is made between the commodity category statistics and the commodity category configuration information, including the commodity categories and corresponding quantities. Each commodity in the commodity category configuration information is checked. If its quantity is not zero, but the quantity of the commodity in the commodity category statistics is zero, the commodity is determined to be out of stock, and out-of-stock information is generated. The commodity category statistics are traversed. If it is found that there are commodity categories that are not listed in the commodity category configuration information, these commodities are determined to be misplaced commodities, and misplaced commodity information is generated. For each commodity in the commodity category statistics, if its quantity is not zero but is less than the quantity specified in the commodity category configuration information, and this quantity is lower than the preset minimum inventory, the commodity is determined to be insufficient in stock, and insufficient inventory information is generated.

[0096] In an example, if the product category configuration information includes 10 product categories a, 3 product categories b, and 4 product categories c, and the product category statistics include 0 product categories a, 1 product category b, 2 product categories c, and 1 product category d, it means that product category a is out of stock, product category d is misplaced, and product categories b and c are out of stock.

[0097] In this embodiment, by comparing the statistical results of commodity categories with the configuration information, out-of-stock, wrong placement and insufficient inventory can be identified more accurately. The automated inventory information matching and related information generation process reduces manual intervention and errors, improves the efficiency and accuracy of commodity management, and reduces operating costs.

[0098] In one embodiment, the commodity booth information includes actual commodity types and actual booth locations in the commodity detection area, and the commodity booth configuration information includes preset commodity types and preset booth locations in the commodity detection area;

[0099] The steps of determining the commodity booth information in the commodity detection area and matching it with the commodity booth configuration information to obtain the commodity inventory information in the commodity detection area include:

[0100] For each preset booth position in the product booth configuration information, if the preset booth position does not have a corresponding actual booth position in the product detection area, out-of-stock information is determined based on the preset booth position and its corresponding preset product type;

[0101] For each preset booth position in the merchandise booth configuration information, if there is a corresponding actual booth position in the merchandise detection area for the preset booth position, and the preset merchandise type corresponding to the preset booth position is different from the actual merchandise type corresponding to the actual booth position, then the merchandise misplacement information is determined based on the preset booth position and its corresponding preset merchandise type to obtain the merchandise inventory information.

[0102] Specifically, first, traverse each preset booth position in the product booth configuration information, and match it with the actual booth position in the product booth information to check whether there is a corresponding relationship. For each preset booth position, if the corresponding actual booth position cannot be found in the corresponding product detection area, that is, no actual product is placed at this position, then the out-of-stock information is determined based on the preset booth position and its corresponding preset product type. If the preset booth position has a corresponding actual booth position in the product detection area, but the product types represented by the two are different, it means that the product is placed in the wrong position. At this time, the product placement error information is determined based on the preset booth position and its corresponding preset product type.

[0103] It is understandable that when determining whether the preset commodity types corresponding to the preset booth location are the same as the actual commodity types corresponding to the actual booth location, barcode / QR code scanning technology, RFID (radio frequency identification) technology, and computer vision and image recognition technology can be used.

[0104] In one example, if the product booth configuration information includes preset product category a+preset booth position box1, preset product category a+preset booth position box2, ..., preset product category a+preset booth position box10, preset product category b+preset booth position box11, preset product category b+preset booth position box12, ..., the product booth information in the product detection area is matched with the product booth configuration information. If box1 is not matched, it is out of stock. If box2 is matched, but the product type is d, it is wrong.

[0105] In this embodiment, by matching the preset and actual booth locations, out-of-stock and placement errors of goods can be discovered in a timely manner. The automated matching and information generation process reduces manual intervention and errors, and improves work efficiency and accuracy.

[0106] In one embodiment, the process of determining that the preset commodity type corresponding to the preset booth position is different from the actual commodity type corresponding to the actual booth position includes:

[0107] Determine a preset product image corresponding to the preset booth position, and determine an actual product image corresponding to the actual booth position;

[0108] An IOU value between the preset product image and the actual product image is calculated. When the IOU value is less than a preset threshold, the preset product type corresponding to the preset booth position is different from the actual product type corresponding to the actual booth position.

[0109] Among them, in image processing and computer vision, IOU (Intersection over Union) is an indicator to measure the degree of overlap between two areas, which calculates the ratio of the intersection area to the union area of ​​the two areas. In product recognition, the IOU value can be used to measure the similarity or overlap between the preset product image and the actual product image. The preset threshold is a pre-set IOU value limit used to determine whether the preset product image is similar enough to the actual product image. When the IOU value is less than this preset threshold, it is considered that the difference between the two is large, that is, the types of goods at the preset booth location and the actual booth location are different.

[0110] Specifically, first, the preset product images corresponding to the preset booth positions can be obtained by means of database query or image library retrieval, and at the same time, the actual product images corresponding to the actual booth positions can be obtained in real time using image capture devices such as cameras. After that, necessary preprocessing operations can be performed on the obtained images, such as size adjustment, noise removal, color correction, etc., to ensure that the image quality meets the requirements of subsequent processing.

[0111] Using image processing algorithms or deep learning models, extract key features from the preprocessed image, which can be color, texture, shape, etc., for subsequent similarity calculations. Calculate the overlapping area of ​​the preset product image and the actual product image, and find their intersection area and union area. Then, according to the definition of the IOU value, calculate the ratio of the intersection area to the union area to obtain the IOU value. Compare the calculated IOU value with the preset threshold. If the IOU value is less than the preset threshold, it is considered that the preset product type corresponding to the preset booth position is different from the actual product type corresponding to the actual booth position; otherwise, the two are considered to be the same or have a high degree of similarity.

[0112] In this embodiment, by calculating the IOU value and comparing it with the preset threshold, it is possible to more accurately determine whether the types of goods at the preset booth location and the actual booth location are the same. Since the IOU value takes into account the degree of overlap of the area, even if there are certain differences between the two goods in color, shape, etc., as long as their core areas can overlap well, they can be correctly identified as the same or similar types of goods, reducing misjudgments caused by local differences. At the same time, the IOU calculation does not rely on specific features such as product shape, color or texture, but is based on the overall area of ​​the image for identification. Therefore, it can be applied to the identification of a variety of different shapes, colors and textures of goods, and has strong adaptability and flexibility. The entire recognition process can be completed automatically by computer without human intervention. This can not only improve work efficiency, but also reduce the impact of human errors and subjective factors.

[0113] In one embodiment, the commodity booth information includes actual commodity types and their actual booth order in the commodity detection area, and the commodity booth configuration information includes preset commodity types and their preset booth order in the commodity detection area;

[0114] The steps of determining the commodity booth information in the commodity detection area and matching it with the commodity booth configuration information to obtain the commodity inventory information in the commodity detection area include:

[0115] Determine a first sequence corresponding to the commodity booth information according to actual commodity types and actual booth sequence in the commodity detection area;

[0116] Determine a second sequence corresponding to the commodity booth configuration information according to the preset commodity types and the preset booth sequence in the commodity detection area;

[0117] Find the longest common subsequence between the first sequence and the second sequence;

[0118] Compare the longest common subsequence with the second sequence, obtain the commodity types that exist in the second sequence but not in the longest common subsequence, and determine out-of-stock information;

[0119] The longest common subsequence is compared with the first sequence to obtain the commodity types that exist in the first sequence but not in the longest common subsequence, and the commodity misplacement information is determined to obtain the commodity inventory information.

[0120] Specifically, according to the actual commodity types in the commodity detection area and the actual booth order, an ordered list is constructed, in which each element represents the commodity type on a booth, and the order is consistent with the actual booth order. According to the preset commodity types in the commodity detection area and the preset booth order, an ordered list is also constructed, in which each element represents the commodity type that should be placed in each booth in the preset, and the order is consistent with the preset booth order. Use dynamic programming or other algorithms to calculate the longest common subsequence between the first sequence and the second sequence. The longest common subsequence refers to the longest subsequence shared by two sequences and arranged in the order of the original sequence.

[0121] Compare the second sequence with the longest common subsequence obtained. For the types of goods in the second sequence that exist in the preset booth but are not in the longest common subsequence, they are regarded as out-of-stock goods, and their out-of-stock information is recorded. These goods are supposed to exist in the preset but are actually missing. Compare the first sequence with the longest common subsequence obtained. For the types of goods in the first sequence that exist in the actual booth but are not in the longest common subsequence, they are regarded as incorrectly placed goods, and their incorrect placement information is recorded. Although these goods exist in the actual booth, they are not placed according to the preset booth order or type.

[0122] In an example, if the sequence corresponding to the product booth configuration information is: a1, a2, b3, b4, c5, d6, e7, f8, g9, and the sequence corresponding to the product booth information is: a1, b2, a3, c4, d5, h6, f7, g8, the two sequences are matched, and the longest common subsequence matched is: aacdfg or abcdfg. One of the two results is randomly matched. If it is the former aacdfg: it means that b3, b4, and e7 are out of stock, and h6 is placed incorrectly; if it is the latter abcdfg, it means that a1 or a2, b3 or b4, e7 are out of stock, and a3 and h6 are placed incorrectly.

[0123] It is understandable that most algorithms for calculating LCS (Longest Common Subsequence) (such as dynamic programming) are designed to find and return an LCS when implemented. These algorithms usually construct a sequence in LCS through a backtracking process, rather than constructing all possible LCSs at the same time. Therefore, from the perspective of algorithm implementation, the output result is naturally an LCS. Based on this, applying this LCS to product booth matching will ultimately lead to only one conclusion.

[0124] In this embodiment, compared with simple category matching or position matching, the longest common subsequence method can more comprehensively consider multiple dimensions (category and order) of product booth information, thereby reducing misjudgments caused by inaccurate matching of a single dimension. At the same time, it does not rely on specific product categories or booth layouts. As long as the corresponding sequence can be constructed, matching and identification can be performed. Therefore, it has strong adaptability and flexibility and can be applied to product detection areas of different sizes and layouts. It can be understood that the calculation of the longest common subsequence can be completed in a shorter time, thereby improving the efficiency of the entire matching process. In addition, since the need for misjudgments and repeated inspections is reduced, the overall work efficiency is also indirectly improved.

[0125] The following is a description of the commodity management device provided in the embodiment of the present application. The commodity management device described below and the commodity management method described above can be referred to in correspondence with each other. Figure 2 As shown, the present application provides a commodity management device, the device comprising:

[0126] The commodity detection strategy determination module 201 is used to determine the commodity detection strategy and the corresponding commodity detection area of ​​the commodity management request in response to the commodity management request;

[0127] The commodity category statistical result determination module 202 is used to obtain commodity category configuration information of the commodity detection area if the commodity detection strategy includes a commodity category matching method, and to count the commodity category information in the commodity detection area to obtain commodity category statistical results;

[0128] The first commodity inventory information determination module 203 is used to match the commodity category statistics result with the commodity category configuration information to obtain the commodity inventory information of the commodity detection area, and the commodity inventory information includes commodity misplacement information and out-of-stock information;

[0129] The commodity booth configuration information acquisition module 204 is used to acquire the commodity booth configuration information of the commodity detection area if the commodity detection strategy includes the commodity booth matching mode;

[0130] The second commodity inventory information determination module 205 is used to determine the commodity booth information in the commodity detection area and match it with the commodity booth configuration information to obtain the commodity inventory information of the commodity detection area.

[0131] In one embodiment, the commodity category statistical result determination module 202 includes:

[0132] A commodity image determination unit, used to crop the image of the commodity detection area to obtain multiple commodity images in the commodity detection area;

[0133] A commodity feature determination unit, used to input each commodity image into a predetermined feature extraction model to obtain commodity features of each commodity image;

[0134] The commodity category statistical result determination unit is used to determine the commodity category corresponding to each commodity in the commodity detection area according to the characteristics of each commodity, so as to obtain the commodity category statistical result.

[0135] In one embodiment, if the commodity category statistics result only includes commodity categories in the commodity detection area, the first commodity inventory information determination module 203 includes:

[0136] A first category matching unit, used to match the commodity category statistical result with the commodity category configuration information;

[0137] A first commodity misplacement information determining unit, configured to determine commodity misplacement information according to the commodity type if there is a commodity type different from the commodity type configuration information in the commodity type statistical result;

[0138] The first out-of-stock information determining unit is used to determine out-of-stock information according to the commodity category if the commodity category in the commodity category configuration information is missing in the commodity category statistical result, so as to obtain commodity inventory information.

[0139] In one embodiment, the commodity inventory information further includes insufficient inventory information;

[0140] If the commodity category statistics result includes commodity categories and corresponding commodity quantities in the commodity detection area, the first commodity inventory information determination module 203 includes:

[0141] A second category matching unit, used to match the commodity category statistical result with the commodity category configuration information;

[0142] a second out-of-stock information determining unit, configured to determine out-of-stock information according to a commodity category if there is a commodity category whose quantity is not zero in the commodity category configuration information and the commodity quantity of the commodity category in the commodity category statistical result is zero;

[0143] A second commodity misplacement information determining unit, configured to determine commodity misplacement information according to the commodity category if the commodity category statistical result includes a commodity category not listed in the commodity category configuration information;

[0144] The insufficient inventory information determining unit is used to determine the insufficient inventory information according to each commodity category in the commodity category statistical result to obtain commodity inventory information if the commodity quantity of the commodity category is not zero and is less than the commodity quantity of the commodity category in the commodity category configuration information.

[0145] In one embodiment, the commodity booth information includes actual commodity types and actual booth locations in the commodity detection area, and the commodity booth configuration information includes preset commodity types and preset booth locations in the commodity detection area;

[0146] The second commodity inventory information determination module 205 includes:

[0147] A third out-of-stock information determining unit is configured to determine out-of-stock information for each preset booth position in the commodity booth configuration information according to the preset booth position and its corresponding preset commodity type if the preset booth position does not have a corresponding actual booth position in the commodity detection area;

[0148] The third unit for determining the misplaced merchandise information is used to determine, for each preset booth position in the merchandise booth configuration information, the misplaced merchandise information based on the preset booth position and its corresponding preset merchandise type, to obtain merchandise inventory information if the preset booth position has a corresponding actual booth position in the merchandise detection area, and the preset merchandise type corresponding to the preset booth position is different from the actual merchandise type corresponding to the actual booth position.

[0149] In one embodiment, the third commodity misplacement information determining unit includes:

[0150] A commodity image determination subunit, used to determine a preset commodity image corresponding to the preset booth position, and to determine an actual commodity image corresponding to the actual booth position;

[0151] The commodity category recognition result determination subunit is used to calculate the IOU value between the preset commodity image and the actual commodity image. When the IOU value is less than a preset threshold, the preset commodity category corresponding to the preset booth position is different from the actual commodity category corresponding to the actual booth position.

[0152] In one embodiment, the commodity booth information includes actual commodity types and their actual booth order in the commodity detection area, and the commodity booth configuration information includes preset commodity types and their preset booth order in the commodity detection area;

[0153] The second commodity inventory information determination module 205 includes:

[0154] A first sequence determination unit, configured to determine a first sequence corresponding to the commodity booth information according to actual commodity types and actual booth sequence in the commodity detection area;

[0155] A second sequence determination unit, configured to determine a second sequence corresponding to the commodity booth configuration information according to the preset commodity types and the preset booth sequence in the commodity detection area;

[0156] A longest common subsequence determination unit, used to obtain the longest common subsequence of the first sequence and the second sequence;

[0157] a fourth out-of-stock information determining unit, configured to compare the longest common subsequence with the second sequence, obtain commodity types that exist in the second sequence but not in the longest common subsequence, and determine out-of-stock information;

[0158] The fourth commodity misplacement information determining unit is used to compare the longest common subsequence with the first sequence, obtain commodity types that exist in the first sequence but not in the longest common subsequence, and determine commodity misplacement information to obtain commodity inventory information.

[0159] In one embodiment, the present application further provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the commodity management method as described in any one of the above embodiments.

[0160] In one embodiment, the present application further provides a computer device having computer-readable instructions stored therein. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the commodity management method as described in any one of the above embodiments.

[0161] Indicatively, Figure 3 As shown, Figure 3 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. The computer device 300 may be provided as a server. Figure 3 The computer device 300 includes a processing component 302, which further includes one or more processors, and a memory resource represented by a memory 301, for storing instructions that can be executed by the processing component 302, such as an application. The application stored in the memory 301 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 302 is configured to execute instructions to perform the commodity management method of any of the above embodiments.

[0162] The computer device 300 may further include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 may operate based on an operating system stored in the memory 301, such as Windows Server TM, Mac OS X TM, Unix TM, Linux TM, Free BSD TM, or the like.

[0163] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0164] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not clearly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of more restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or equipment including the elements. Herein, "one", "one", "said", "the" and "it" may also include plural forms, unless the context clearly indicates another way. A plurality refers to at least two cases, such as 2, 3, 5 or 8, etc. "And / or" includes any and all combinations of the relevant listed items.

[0165] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can refer to each other.

[0166] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A commodity management method, It is characterized in that The method comprises: In response to a commodity management request, determining a commodity detection strategy and a corresponding commodity detection area of ​​the commodity management request; If the commodity detection strategy includes a commodity category matching method, then obtaining commodity category configuration information of the commodity detection area, and counting commodity category information in the commodity detection area to obtain commodity category statistics results; Matching the commodity category statistics result with the commodity category configuration information to obtain commodity inventory information of the commodity detection area, wherein the commodity inventory information includes commodity misplacement information and out-of-stock information; If the commodity detection strategy includes a commodity booth matching method, obtaining commodity booth configuration information of the commodity detection area; The commodity booth information in the commodity detection area is determined and matched with the commodity booth configuration information to obtain the commodity inventory information of the commodity detection area.

2. The commodity management method according to claim 1, It is characterized in that The step of counting commodity category information in the commodity detection area to obtain commodity category statistics results includes: Cropping the image of the commodity detection area to obtain a plurality of commodity images in the commodity detection area; Inputting each of the product images into a predetermined feature extraction model to obtain product features of each of the product images; The commodity category corresponding to each commodity in the commodity detection area is determined according to each of the commodity characteristics to obtain the commodity category statistical result.

3. The commodity management method according to claim 1, It is characterized in that If the commodity category statistical result only includes commodity categories in the commodity detection area, the step of matching the commodity category statistical result with the commodity category configuration information to obtain commodity inventory information in the commodity detection area includes: Matching the commodity category statistical result with the commodity category configuration information; If there is a commodity category different from the commodity category configuration information in the commodity category statistical result, determining commodity misplacement information according to the commodity category; If the commodity category statistics result lacks the commodity category in the commodity category configuration information, out-of-stock information is determined based on the commodity category to obtain the commodity inventory information.

4. The commodity management method according to claim 1, It is characterized in that The commodity inventory information also includes insufficient inventory information; If the commodity type statistical result includes the commodity types and the corresponding commodity quantities in the commodity detection area, the step of matching the commodity type statistical result with the commodity type configuration information to obtain out-of-stock information and commodity misplacement information in the commodity detection area includes: Matching the commodity category statistical result with the commodity category configuration information; If there is a commodity category whose commodity quantity is not zero in the commodity category configuration information, and the commodity quantity of the commodity category in the commodity category statistical result is zero, out-of-stock information is determined based on the commodity category; If the commodity category statistics result includes commodity categories that are not listed in the commodity category configuration information, determining commodity misplacement information based on the commodity category; If for each commodity category in the commodity category statistics, the commodity quantity of the commodity category is not zero and is less than the commodity quantity of the commodity category in the commodity category configuration information, then insufficient inventory information is determined based on the commodity category to obtain the commodity inventory information.

5. The commodity management method according to claim 1, It is characterized in that The commodity booth information includes actual commodity types and actual booth locations in the commodity detection area, and the commodity booth configuration information includes preset commodity types and preset booth locations in the commodity detection area; The step of determining the commodity booth information in the commodity detection area and matching it with the commodity booth configuration information to obtain the commodity inventory information in the commodity detection area includes: For each preset booth position in the commodity booth configuration information, if the preset booth position does not have a corresponding actual booth position in the commodity detection area, determining out-of-stock information according to the preset booth position and its corresponding preset commodity type; For each preset booth position in the merchandise booth configuration information, if the preset booth position has a corresponding actual booth position in the merchandise detection area, and the preset merchandise type corresponding to the preset booth position is different from the actual merchandise type corresponding to the actual booth position, then the merchandise misplacement information is determined based on the preset booth position and its corresponding preset merchandise type to obtain the merchandise inventory information.

6. The commodity management method according to claim 5, It is characterized in that The process of determining that the preset commodity type corresponding to the preset booth position is different from the actual commodity type corresponding to the actual booth position includes: Determine a preset product image corresponding to the preset booth position, and determine an actual product image corresponding to the actual booth position; An IOU value between the preset product image and the actual product image is calculated. When the IOU value is less than a preset threshold, the preset product type corresponding to the preset booth position is different from the actual product type corresponding to the actual booth position.

7. The commodity management method according to claim 1, It is characterized in that The commodity booth information includes actual commodity types and actual booth order in the commodity detection area, and the commodity booth configuration information includes preset commodity types and preset booth order in the commodity detection area; The step of determining the commodity booth information in the commodity detection area and matching it with the commodity booth configuration information to obtain the commodity inventory information in the commodity detection area includes: Determine a first sequence corresponding to the commodity booth information according to actual commodity types and actual booth order in the commodity detection area; Determining a second sequence corresponding to the commodity booth configuration information according to the preset commodity types and the preset booth sequence in the commodity detection area; Finding the longest common subsequence of the first sequence and the second sequence; Compare the longest common subsequence with the second sequence to obtain commodity types that exist in the second sequence but not in the longest common subsequence, and determine out-of-stock information; The longest common subsequence is compared with the first sequence to obtain commodity types that exist in the first sequence but not in the longest common subsequence, and commodity misplacement information is determined to obtain the commodity inventory information.

8. A commodity management device, It is characterized in that The device comprises: A commodity detection strategy determination module, configured to determine, in response to a commodity management request, a commodity detection strategy and a corresponding commodity detection area of ​​the commodity management request; A commodity category statistical result determination module, configured to obtain commodity category configuration information of the commodity detection area if the commodity detection strategy includes a commodity category matching method, and to count commodity category information in the commodity detection area to obtain commodity category statistical results; A first commodity inventory information determination module, used for matching the commodity category statistics result with the commodity category configuration information to obtain commodity inventory information of the commodity detection area, wherein the commodity inventory information includes commodity misplacement information and out-of-stock information; A commodity booth configuration information acquisition module, configured to acquire the commodity booth configuration information of the commodity detection area if the commodity detection strategy includes a commodity booth matching method; The second commodity inventory information determination module is used to determine the commodity booth information in the commodity detection area and match it with the commodity booth configuration information to obtain the commodity inventory information of the commodity detection area.

9. A storage medium, Features: The storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the commodity management method according to any one of claims 1 to 7.

10. A computer device, It is characterized in that include: one or more processors, and memory; The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the one or more processors, the steps of the commodity management method according to any one of claims 1 to 7 are performed.

Citation Information

Cited By

  • Goods shelf management method and system based on target detection, terminal and medium

    CN119784995A