Product image information collection system and method

Abstract

To provide a system and a method for commodity image information collection which improve the efficiency of collecting images and information of commodities.SOLUTION: A commodity image information collection system comprises: an AIoT management server which transmits a commodity information list to a price tag management server and transmits an image capture start notification to an AI vision system; a price tag management server which, on the basis of binding information between electronic price tags and commodities and the received commodity information list, determines an electronic price tag to be notified, transmits an information display notification, and, when the electronic price tag to be notified performs an information display operation based on the notification, simultaneously generates a matching relationship between the information display operation and a barcode of the commodity and transmits the matching relationship to the AI vision system; an AI vision system which captures an image of the information display operation to obtain captured content, and performs parsing on the basis of the captured content and the matching relationship to obtain a recognition result; and a commodity information management system which associates the electronic price tag with the commodity on the basis of the captured image and the recognition result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the technical field of mobile interconnection, and in particular to a product image information collection system and method. [Background technology]

[0002] This section is intended to provide a background or context for the claimed embodiments of the invention. Nothing herein is admitted as prior art for inclusion in this section.

[0003] With the continuous development of artificial intelligence technology, the supermarket industry is also moving toward intelligence and digitalization. By applying AI technology, supermarkets can better meet customer needs, improve operational efficiency, enhance user experience, and further expand their market influence. By applying AI algorithms, supermarkets can analyze and predict consumer needs in real time, more precisely formulating marketing strategies to meet consumer needs, leading to increased sales and customer satisfaction. Supermarkets can also use AI technology to automate inventory management, reducing inventory costs and shelf stagnation rates, and quickly replenish inventory when needed to meet customer shopping needs.

[0004] Smart shelves are a new type of smart retail tool that features smart analysis and automated management functions and can meet various needs such as supermarket display analysis and product recognition. They generally include hardware devices such as electronic price tags, smart cameras, and smart robots, and perform real-time monitoring and data analysis of products on shelves, thereby improving shelf utilization, optimizing inventory management, improving customer shopping experience, and increasing supermarket operating profits. Smart shelves have already become an important component of smart retail in supermarkets and will also be an important direction for the future development of supermarkets.

[0005] Collecting product image information is an important part of smart shelving. The collected product image data can be used for subsequent tasks such as product recognition, display analysis, and out-of-stock checks. These tasks require not only collected product images, but also accurate corresponding barcodes and binding of product information to images.

[0006] However, the collection of product data with similar specifications is usually done manually, using dedicated collection equipment to scan products with a code scanning gun to obtain barcodes, and then using a few cameras to capture images of the products for collection and processing. This method involves the transportation and movement of products, and not only is the collection efficiency very low, but new products are often introduced in supermarkets, making it very difficult to collect data using dedicated equipment. Summary of the Invention [Problem to be solved by the invention]

[0007] An embodiment of the present invention provides a product image information collection system configured to improve the efficiency of collecting image and information of products, particularly additional products or specific products. [Means for solving the problem]

[0008] The product image information collection system according to the embodiment of the present invention includes: an AIoT management server configured to send a product information list to the price tag management server and send an imaging start notification to the AI vision system; a price tag management server configured to determine an electronic price tag to be notified based on binding information between the electronic price tag and the product and the received product information list, to send an information display notification to the electronic price tag to be notified, to have the electronic price tag to be notified perform an information display action based on the information display notification, and at the same time to generate a matching relationship between the information display action and the product barcode and send it to the AI vision system; an AI vision system configured to, upon receiving an image capture start notification, capture an image of the information display action to obtain the image content, analyze the image content and the received matching relationship to obtain a recognition result including spatial information and recognition information of the electronic price tag and the product, and transmit the captured image and the recognition result to a product information management system; and a product information management system configured to associate electronic price tags with products based on the captured images and the recognition results.

[0009] An embodiment of the present invention provides a commodity image information collection method adapted to an AI vision system, and configured to improve efficiency of collecting commodity image information, particularly additional commodity or specific commodity image information, the commodity image information collection method comprising: When receiving an imaging start notification transmitted by the AIoT management server, imaging the information exhibition operation and acquiring the imaging content; A step of analyzing the image content and a matching relationship, which is the relationship between the information display action transmitted by the price tag management server and the barcode of the product, to obtain a recognition result including spatial information and recognition information of the electronic price tag and the product, wherein the matching relationship is generated at the same time that the electronic price tag to be notified by the price tag management server executes the information display action based on the information display notification, and the information display notification is transmitted to the electronic price tag to be notified after the price tag management server determines the electronic price tag to be notified based on the binding information between the electronic price tag and the product and the product information list transmitted by the received AIoT management server; and transmitting the captured image and the recognition result to the product information management system so that the product information management system associates the electronic price tag with the product.

[0010] An embodiment of the present invention provides a product image information collection device that is applied to an AI vision system and configured to improve efficiency of collecting image and information of a product, particularly an additional product or a specific product, and the product image information collection device includes: An imaging module configured to capture an image of the information exhibition operation and acquire the captured image content upon receiving an imaging start notification transmitted by the AIoT management server; A recognition module configured to analyze the image content and a matching relationship between the information display action transmitted by the price tag management server and the barcode of the product to obtain a recognition result including spatial information and recognition information of the electronic price tag and the product, wherein the matching relationship is generated at the same time that the electronic price tag to be notified by the price tag management server executes the information display action based on the information display notification, and the information display notification is transmitted by the price tag management server to the electronic price tag to be notified after the electronic price tag to be notified is determined based on the binding information between the electronic price tag and the product and the product information list transmitted by the received AIoT management server; and an association module configured to transmit the captured image and the recognition result to the product information management system so that the electronic price tag and the product are associated by the product information management system.

[0011] In an embodiment of the present invention, the AIoT management server is configured to send a product information list to the price tag management server and send an imaging start notification to the AI vision system. The price tag management server is configured to determine the electronic price tag to be notified based on the binding information between the electronic price tag and the product and the received product information list, send an information display notification to the electronic price tag to be notified, and while the electronic price tag to be notified performs an information display operation based on the information display notification, generate a matching relationship between the information display operation and the product barcode and send it to the AI vision system. Upon receiving the imaging start notification, the AI vision system is configured to image the information display operation to obtain the image content, and analyze it based on the image content and the received matching relationship to obtain a recognition result including spatial information and recognition information of the electronic price tag and the product, and send the imaged image and the recognition result to the product information management system. The product information management system is configured to associate the electronic price tag with the product based on the imaged image and the recognition result. Compared with the manual collection method used in conventional technology, this technical solution uses AIoT technology to build the entire system, which performs automatic discovery, automatic control, automatic recognition, and automatic analysis, improving the efficiency of collecting product images and information, and efficiently performing multiple rounds of collection, thereby reducing the cost of repeated manual collection and achieving a closed-loop data loop in smart shelf operations. [Brief explanation of the drawings]

[0012] In order to more clearly describe the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces drawings necessary for the description of the embodiments or the prior art, obviously, the drawings in the following description are only some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without inventive work.

[0013] [Figure 1] 1 is a schematic diagram of a product image information collection system according to an embodiment of the present invention. [Figure 2] FIG. 2 is another schematic diagram of a product image information collection system according to an embodiment of the present invention. [Figure 3]FIG. 10 is a diagram illustrating a calculation principle of the similarity between an anchor product and other products according to an embodiment of the present invention. [Figure 4] 2 is a flowchart of a product image information collecting method according to an embodiment of the present invention. [Figure 5] 1 is a schematic diagram of a product image information collection device according to an embodiment of the present invention. [Figure 6] 1 is a schematic diagram of a computer device according to an embodiment of the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0014] In order to make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in more detail below with reference to the drawings. Hereinafter, the exemplary embodiments of the present invention and the description thereof are for the purpose of interpreting the present invention, but are not intended to limit the present invention.

[0015] The principle of the embodiment of the present invention is to adopt digitalization-related equipment deployed in supermarkets, and based on AIoT technology, automatically perform imaging, recognition inference, calculation analysis, and storage, thereby achieving the goals of automatic discovery, automatic control, automatic recognition, automatic analysis, and automatic storage in a base, improving the efficiency of collecting product images and information, and enabling multiple rounds of collection to be efficiently performed, reducing the cost of repeated manual collection, and achieving a data closed loop in smart shelf operations.

[0016] The electronic price tag in the embodiment of the present invention refers to the electronic product price tag commonly seen in conventional supermarkets, which can be turned into pages and has an attached LED lamp that can flash.

[0017] FIG. 1 is a schematic diagram of a product image information collection system according to an embodiment of the present invention; an AIoT management server configured to send a product information list to the price tag management server and send an imaging start notification to the AI vision system; a price tag management server configured to determine an electronic price tag to be notified based on binding information between the electronic price tag and the product and the received product information list, to send an information display notification to the electronic price tag to be notified, to have the electronic price tag to be notified perform an information display action based on the information display notification, and at the same time to generate a matching relationship between the information display action and the product barcode and send it to the AI vision system; an AI vision system configured to, upon receiving an image capture start notification, capture an image of the information display action to obtain the image content, analyze the image content and the received matching relationship to obtain a recognition result including spatial information and recognition information of the electronic price tag and the product, and transmit the captured image and the recognition result to a product information management system; and a product information management system configured to associate electronic price tags with products based on the captured images and the recognition results.

[0018] Referring to FIG. 2, the system includes: Further, the code scanning device is configured to transmit binding information between the electronic price tags and the products on the shelf to the price tag management server; The price tag management server is further configured to record binding information between the electronic price tag and the product and provide it to the AIoT management server in synchronization.

[0019] On a daily basis, supermarket staff use code scanning devices to bind electronic price tags to products on the shelves.

[0020] In one embodiment, the AIoT management server further comprises: Calculate the statistics of the added products, obtain the added product information from the price tag management server, import the added product information, and add the added product information to the product information list; Alternatively, the device is configured to receive a preset product information list.

[0021] The above are two scenarios for obtaining a product information list. In these two scenarios, the AI vision system can be configured to capture images in several rounds, thereby acquiring more product images for subsequent steps. The additional product information can be from the price tag management server, or can be provided to the AIoT management server in an externally introduced manner, or can be provided in other ways, and this is not limited thereto.

[0022] In one embodiment, the information display action is a page change or a blinking action; If the information display operation is page switching, the captured content is a content image of the page switching; When the information display action is blinking, the captured content is a sequence of blinking images.

[0023] In a specific implementation, the AI vision system controls a visual sensor device to capture images and perform AI recognition and inference calculations on the visual images. Visual sensor devices include, but are not limited to, fixedly mounted cameras, movable cameras, rotatable cameras, mobile camera-equipped robots, and camera-equipped aircraft. These visual sensor devices capture images of supermarket shelves by looking down, looking up, or directly at them. When the information display behavior is a page change, a specific page change code is displayed for recognition by the camera. When the information display behavior is a blinking display, the blinking image sequence is captured by capturing the blinking.

[0024] In one embodiment, the spatial information is a position bounding box; Specifically, the AI vision system: Discard the image content without information display operation, Detecting EPL tags in the remaining image content and obtaining a position bounding box for each EPL tag; For each detected electronic price tag, the information display action is recognized, Analyzing the information display action based on a matching relationship between the information display action and the barcode of the product, and determining the barcode of the product corresponding to each electronic price tag; The system is configured to detect items in the image content and obtain a position bounding box for each item.

[0025] Through the above AI calculation, the position bounding box of the EPL and the position bounding box of the product are obtained, and the barcode of the product corresponding to each EPL is also obtained, and this information is used for subsequent association. Methods for detecting EPL and products include, but are not limited to, conventional visual processing algorithms and deep learning methods. When using deep learning methods, target detection algorithms pre-trained by machine learning methods, image segmentation algorithms, etc. are used. Methods for recognizing information display behavior (i.e., page switching content or flashing of the EPL) similarly include, but are not limited to, conventional visual processing algorithms and deep learning methods.

[0026] In one embodiment, the AI vision system recognizes the information display action by: For each detected EPL, cropping a small image of the EPL based on the position bounding box of the EPL; Recognizing each price tag sub-image and recognizing an information display action.

[0027] The step of extracting the small price tag image described above can accelerate the recognition of the information display action.

[0028] In one embodiment, the product information management system specifically includes: determining a product closest to each electronic price tag as an anchor product corresponding to the electronic price tag based on the spatial information of the electronic price tag and the spatial information of the product; calculate a similarity between the image content of each other product and the image content of the anchor product within a predetermined range for each anchor product, and if the similarity corresponding to each other product is equal to or greater than a predetermined threshold, determine that the other product and the anchor product are of the same type of product; Each electronic price tag and corresponding anchor product are configured to be associated and bound with other products that are the same type of product as the anchor product.

[0029] In concrete implementation, the other products within a predetermined range of the anchor product refer to the n products stacked on either side of or on top of the anchor product. Associating and binding each electronic price tag and its corresponding anchor product with other products of the same type as the anchor product includes binding the products with information, thereby providing each associated product with clear barcode information, product name information, etc.

[0030] FIG. 3 illustrates the principle of calculating the similarity between an anchor product and other products according to an embodiment of the present invention. For electronic price tag A, an anchor product is determined based on the closest distance relationship, and the similarity between the anchor product and other surrounding products is calculated. The number displayed on the head of each other product represents the image similarity s with the anchor product. If the threshold is 0.85, three other products with similarity values of 0.86, 0.91, and 0.95 are considered to be of the same type as the anchor product, form an associated relationship, and can be information-bound with electronic price tag A.

[0031] In one embodiment, the product information management system further comprises: The system is configured to extract a product image from the captured image based on the spatial information of the product, and store the product information corresponding to the product image in a database.

[0032] In one embodiment, the product information management system further comprises: Visualizing and displaying the association binding object including the image content, the recognition result, the anchor product corresponding to the electronic price tag, each other product within a predetermined range, and the association binding product; It is configured to accept manual edits to the object, including additions, deletions, or corrections.

[0033] The above process is carried out in the display interface of the product information management system, that is, the user can intuitively view the association binding, and can also intervene in the association binding to realize adjustment.

[0034] The embodiment of the present invention further proposes a commodity image information collection method applicable to an AI vision system, and the principle thereof is similar to that of the commodity image information collection system, so detailed description thereof will be omitted here.

[0035] FIG. 4 is a flowchart of a commodity image information collection method according to an embodiment of the present invention; Step 401: When receiving the imaging start notification transmitted by the AIoT management server, imaging the information exhibition operation and acquiring the imaging content; Step 402: Analyzing the captured image content and the matching relationship between the information display action and the product barcode transmitted by the price tag management server to obtain a recognition result including spatial information and recognition information of the electronic price tag and the product, wherein the matching relationship is generated at the same time that the electronic price tag to be notified by the price tag management server executes the information display action based on the information display notification, and the information display notification is transmitted to the electronic price tag to be notified after the price tag management server determines the electronic price tag to be notified based on the binding information between the electronic price tag and the product and the product information list transmitted by the received AIoT management server; and step 403 of transmitting the captured image and the recognition result to the product information management system so that the electronic price tag and the product are associated by the product information management system.

[0036] In one embodiment, the step of analyzing and obtaining a recognition result based on the image content and the matching relationship sent by the price tag management server includes: Discarding the image content without the information display action; Detecting EPLs in the remaining image content and obtaining a position bounding box for each EPL; Recognizing an information display action for each detected electronic price tag; Analyzing the information display action based on a matching relationship between the information display action and the barcode of the product, determining the barcode of the product corresponding to each electronic price tag, and setting the determined barcode of the product as recognition information; Detecting items in the image content and obtaining a position bounding box for each item.

[0037] In one embodiment, the step of recognizing the information exhibiting action comprises: For each detected EPL, extracting a small image of the EPL based on the position bounding box of the EPL; and recognizing each small price tag image and recognizing an information display action.

[0038] The embodiment of the present invention further proposes a commodity image information collection device that is applied to an AI vision system, and its principle is similar to that of the commodity image information collection system, so detailed description is omitted here.

[0039] FIG. 5 is a schematic diagram of a product image information collection device according to an embodiment of the present invention; An imaging module 501 configured to capture an image of an information display operation and acquire the captured image content upon receiving an imaging start notification transmitted by the AIoT management server; A recognition module 502 configured to analyze the image content and a matching relationship between the information display action transmitted by the price tag management server and the barcode of the product to obtain a recognition result including spatial information and recognition information of the electronic price tag and the product, wherein the matching relationship is generated at the same time that the electronic price tag to be notified by the price tag management server executes the information display action based on the information display notification, and the information display notification is transmitted to the electronic price tag to be notified after the price tag management server determines the electronic price tag to be notified based on the binding information between the electronic price tag and the product and the product information list transmitted by the received AIoT management server; and and an association module 503 configured to transmit the captured image and the recognition result to the product information management system so that the EPL and the product are associated by the product information management system.

[0040] In one embodiment, the recognition module 502 specifically: Discard the image content without information display operation, Detecting EPL tags in the remaining image content and obtaining a position bounding box for each EPL tag; For each detected electronic price tag, the information display action is recognized, Analyzing the information display action based on a matching relationship between the information display action and the barcode of the product, and determining the barcode of the product corresponding to each electronic price tag; The system is configured to detect items in the image content and obtain a position bounding box for each item.

[0041] In one embodiment, the recognition module 502 specifically: For each detected EPL, extract a small image of the EPL based on the position bounding box of the EPL; The device is configured to recognize each small price tag image and recognize information display actions.

[0042] As described above, the system, method and apparatus proposed in the embodiments of the present invention have the following beneficial effects.

[0043] First, the product information management system can automatically calculate, collect and store product images and bound product information based on the information on the images.

[0044] Second, the architecture of the AIoT management system can be simultaneously connected to a price tag management server and an AI vision system, where the price tag management server can be connected to all electronic price tags in the scene, and the AI vision system can be connected to all cameras or devices in the scene, thereby realizing the interconnection of various devices.

[0045] Third, the AIoT management system can detect additional products based on database information and send the list of added products or the configured product list to the price tag management server, thereby controlling the page flipping or flashing of electronic price tags. This allows for quick acquisition of image information of manually added and highlighted products in supermarket scenes, improving work efficiency.

[0046] Fourth, the AI vision system works synchronously under the notification of the AIoT management server, and can acquire the corresponding image sets when the price tag changes pages or flashes, and perform inference calculations, eliminating the need for repeated manual collection and avoiding low efficiency and high costs.

[0047] Fifth, the AI vision system can detect electronic price tags and products in shelf images, and can also recognize the content of page changes or the color of flashing lights on electronic price tags.

[0048] An embodiment of the present invention further provides a computer device. FIG. 6 is a schematic diagram of a computer device according to an embodiment of the present invention. The computer device 600 includes a memory 610, a processor 620, and a computer program 630 stored in the memory 610 and executable by the processor 620. When the computer program 630 is executed by the processor 620, the above-mentioned product image information collection method is realized.

[0049] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, realizes the above-described product image information collection method.

[0050] An embodiment of the present invention further provides a computer program that, when executed by a processor, realizes the above-described product image information collection method.

[0051] Those skilled in the art should understand that embodiments of the present invention may be provided as a method, a system, or a computer program product. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. The present invention may also take the form of a computer program product embodied in one or more computer-usable storage media (including, but not limited to, magnetic disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0052] The present invention has been described with reference to flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block of the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or a processor of other programmable data processing devices to generate an apparatus for implementing the functions specified in one or more flows of the flowcharts and / or one or more blocks of the block diagrams, such that the instructions, executed by the processor of the computer or other programmable data processing devices, generate an apparatus.

[0053] These computer program instructions may be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, thereby producing an article of manufacture including an instruction apparatus that implements the functions specified in one or more flows of the flowcharts and / or one or more blocks of the block diagrams with the instructions stored in the computer-readable memory.

[0054] These computer program instructions can be loaded into a computer or other programmable data processing device, thereby generating a computer-implemented process in steps that perform a series of operations on the computer or other programmable device, thereby providing steps for implementing the functions specified in one or more flows of the flowcharts and / or one or more blocks of the block diagrams, by the instructions that execute on the computer or other programmable device.

[0055] The specific embodiments described above have further explained the objectives, technical solutions and beneficial effects of the present invention, but it should be understood that the above are only specific embodiments of the present invention and are not intended to limit the scope of the present invention. Any amendments, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention should all be included within the scope of the present invention.

Claims

1. An AIOT management server configured to send a product information list to the price tag management server and send an imaging start notification to the AI vision system; a price tag management server configured to determine the electronic price tag to be notified based on binding information between the electronic price tag and the product and the received product information list, to send an information display notification to the electronic price tag to be notified, to have the electronic price tag to be notified perform an information display action based on the information display notification, and at the same time to generate a matching relationship between the information display action and the product barcode and send it to the AI vision system; the AI vision system is configured to, upon receiving an imaging start notification, capture an image of the information display action to obtain the image content, and if the information display action is captured in the image content, analyze the image content based on the image content and the received matching relationship, recognize spatial information of the electronic price tag in the image content, spatial information of the product, and identification information which is the barcode of the product, obtain a recognition result including the spatial information of the electronic price tag, the spatial information of the product, and the identification information of the product, and transmit the image content and the recognition result to a product information management system; and a product information management system configured to determine a plurality of related products of the electronic price tag based on the image content and the recognition result, and to assign information of the products bound to the electronic price tag to the plurality of related products. A product image information collection system characterized by:

2. The method further includes a code scanning device configured to transmit the binding information between the electronic price tag and the product on the shelf to the price tag management server; The price tag management server is further configured to record the binding information between the electronic price tag and the product.

2. The product image information collection system according to claim 1.

3. The AIot management server further The system is configured to collect statistics on additional products, obtain additional product information from the price tag management server, import the additional product information, and add the additional product information to the product information list.

2. The product image information collection system according to claim 1.

4. The information display operation is page switching or blinking, If the information display operation is page switching, the captured content is a content image of the page switching; If the information display action is blinking, the captured image is a sequence of blinking images.

2. The product image information collection system according to claim 1.

5. the spatial information is a position bounding box; The AI vision system Discard the image content without information display operation, Detecting the EPL tags in the remaining image content and obtaining the position bounding box of each EPL tag; Recognizing an information display action for each of the detected electronic price tags; determining a barcode of the product corresponding to each electronic price tag based on the matching relationship between the information display action and the barcode of the product; configured to detect items in the imaged content and obtain the position bounding box for each item.

2. The product image information collection system according to claim 1.

6. The AI visual system recognizes the information display action, For each detected EPL, extracting a small image of the EPL based on the position bounding box of the EPL; and recognizing each price tag sub-image and recognizing an information display action.

6. The product image information collection system according to claim 5.

7. The product information management system includes: determining a product closest to each electronic price tag as an anchor product corresponding to the electronic price tag based on the spatial information of the electronic price tag and the spatial information of the product; The method is configured to calculate the similarity between the image content of each of the other products and the image content of the anchor product within a predetermined range for each of the anchor products, and if the similarity corresponding to each of the other products is equal to or greater than a predetermined threshold, determine that the other products and the anchor product are of the same type of product.

2. The product image information collection system according to claim 1.

8. The product information management system further includes: Based on the spatial information of the product, a product image is extracted from the captured image of the product, and the product information corresponding to the product image is stored in a database.

2. The product image information collection system according to claim 1.

9. The product information management system further includes: Visualizing and displaying the associated binding objects, including the captured image, the recognition result, the anchor product corresponding to the electronic price tag, each of the other products within a predetermined range, and the associated bound products; configured to accept manual edits to the object, including additions, deletions, or corrections; 8. The product image information collection system according to claim 7.

10. In a product image information collection method applied to an AI vision system, When receiving the imaging start notification transmitted by the AIoT management server, imaging the information exhibition operation and acquiring the imaging content; a step of analyzing the imaged content based on the imaged content and a matching relationship between the information display action and the barcode of the product, when the imaged content shows an information display action, recognizing the spatial information of the electronic price tag of the imaged content, the spatial information of the product, and the identification information of the product's barcode, and acquiring a recognition result including the spatial information of the electronic price tag, the spatial information of the product, and the identification information of the product, wherein the matching relationship is generated by the price tag management server at the same time as the electronic price tag to be notified executes the information display action based on the information display notification, and the information display notification is transmitted by the price tag management server to the electronic price tag to be notified after the electronic price tag to be notified is determined based on binding information between the electronic price tag and the product and the product information list received and transmitted by the AIot management server; determining a plurality of related products of the electronic price tag by a product information management system, and transmitting the image content and the recognition result to the product information management system so as to assign product information bound to the electronic price tag to the plurality of related products. A product image information collection method characterized by:

11. In a product image information collection device applied to an AI vision system, an imaging module configured to capture an image of the information exhibition action and acquire the captured image content when receiving an imaging start notification transmitted by the AIoT management server; a recognition module configured to, when an information display action is captured in the captured image, analyze the captured image based on the captured image and a matching relationship between the information display action and the barcode of the product transmitted by a price tag management server, recognize the spatial information of the electronic price tag in the captured image, the spatial information of the product, and the identification information of the product, and acquire a recognition result including the spatial information of the electronic price tag, the spatial information of the product, and the identification information of the product, wherein the matching relationship is generated at the same time that the electronic price tag to be notified by the price tag management server executes the information display action based on an information display notification, and the information display notification is transmitted by the price tag management server to the electronic price tag to be notified after the electronic price tag to be notified is determined based on binding information between the electronic price tag and the product and the product information list received and transmitted by the AIot management server; and an association module configured to determine a plurality of related products of the electronic price tag by a product information management system, and to transmit the image content and the recognition result to the product information management system so as to assign information of the products bound to the electronic price tag to the plurality of related products. A product image information collection device characterized by:

12. A computer device including a memory, a processor, and a computer program stored in the memory and executable by the processor, When the computer program is executed by the processor, the product image information collection method according to claim 10 is realized.

1. A computer device characterized by:

13. A computer program is stored that, when executed by a processor, realizes the product image information collection method according to claim 10. A computer-readable storage medium comprising:

14. A computer program that, when executed by a processor, realizes the product image information collection method according to claim 10.

Images