Convenient store intelligent management method, device and equipment based on multi-modal perception and block chain, and storage medium

Abstract

The invention belongs to the technical field of computers, and discloses a convenience store intelligent management method and device based on multi-modal perception and a block chain, equipment and a storage medium. The method comprises the following steps: identifying a commodity stockout state and a commodity stacking state based on a collected commodity image, and obtaining a real-time inventory based on a commodity weight, the commodity stockout state and the commodity stacking state; classifying the commodities, and associating the commodity identification data with the commodities based on a classification result to generate a block chain traceability code; inputting the real-time inventory into an LSTM model to predict the sales volume, calculating a comprehensive index value according to the commodity stockout state, and removing the to-be-deleted commodities based on the comprehensive index value; and positioning the positions of a plurality of remaining commodities through the block chain traceability code, and triggering an electronic price tag to adjust the price or a stepping motor to adjust the shelf based on the positions so as to realize intelligent management. Through the mode, the stockout state can be sensed in time, replenishment delay is avoided, and meanwhile optimization management can be carried out for classification of commodities.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a convenience store intelligent management method, device, equipment and storage medium based on multimodal perception and blockchain. Background Art

[0002] As a key component of the retail industry, chain convenience stores rely heavily on precise category management and inventory control for operational efficiency. Traditionally, merchandise management relies primarily on manual inventory checks and empirical judgment. However, manual inventory checks are time-consuming and lack real-time detection of out-of-stock and misplaced items, leading to replenishment delays. Furthermore, product elimination relies on static rules, without incorporating dynamic out-of-stock rates, category role weighting, and other comprehensive indicators, resulting in a large number of unsold products. This low level of intelligence leads to replenishment delays and a lack of category optimization in traditional convenience store management models.

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

[0004] The main purpose of the present invention is to provide a convenience store intelligent management method, device, equipment and storage medium based on multimodal perception and blockchain, aiming to solve the technical problems of traditional convenience store management model such as replenishment delays and lack of category optimization.

[0005] To achieve the above objectives, the present invention provides a convenience store intelligent management method based on multimodal perception and blockchain, which includes the following steps:

[0006] The weight sensors, 3D structured light vision sensors and RFID readers deployed on the shelves collect the weight, image and identification data of the goods on the convenience store shelves in real time;

[0007] identifying a product out-of-stock status and a product stacking status based on the product image, and obtaining a real-time inventory quantity based on the product weight, the product out-of-stock status, and the product stacking status;

[0008] Classify the goods into categories, and based on the classification results, associate the product identification data with the goods to generate a blockchain traceability code;

[0009] Input the real-time inventory into the LSTM model to predict sales, calculate a comprehensive index value based on the out-of-stock status of the product, and eliminate the products to be deleted based on the comprehensive index value;

[0010] The locations of the remaining commodities are located through the blockchain traceability code, and electronic price tags are triggered to adjust prices or stepper motors are triggered to adjust shelves based on the locations to achieve intelligent management.

[0011] In some embodiments, obtaining the real-time inventory based on the weight of the product, the out-of-stock status of the product, and the stacking status of the product includes:

[0012] Locating an out-of-stock area based on the out-of-stock status of the product;

[0013] Determining the mass of the missing goods in the located out-of-stock area based on the weight of the goods;

[0014] Determine the effective product quality in the non-out-of-stock area based on the missing product quality;

[0015] Determining an effective commodity density in a non-out-of-stock area based on the commodity stacking status;

[0016] The real-time inventory quantity is determined according to the effective commodity quality and the effective commodity density.

[0017] In some embodiments, determining the effective commodity density of the non-out-of-stock area based on the commodity stacking state includes:

[0018] Call the commodity density database to determine the reference commodity density in the non-out-of-stock area;

[0019] Determining, based on the stacking status of the goods, goods that are blocked by stacking and goods that are not blocked by stacking in the non-out-of-stock area;

[0020] The reference commodity density of the commodities not blocked by stacking is used as the effective commodity density, and the reference commodity density of the commodities blocked by stacking is adjusted by using a preset density compensation coefficient to obtain the effective commodity density.

[0021] In some embodiments, the categorization of goods, associating the goods identification data with the goods based on the categorization results, and generating a blockchain traceability code includes:

[0022] Classify the products into four categories, including target categories, regular categories, convenience categories, and seasonal categories;

[0023] Extracting the weight of each category of goods based on the classification results, and extracting the minimum stock unit and production batch number of the goods from the said product identification data;

[0024] Generate a unique hash value based on the minimum stock keeping unit of the product and the production batch number using the SHA-256 algorithm;

[0025] The hash value is associated with the weight of each category of goods, and is written into the blockchain through the PBFT consensus mechanism to obtain a blockchain traceability code.

[0026] In some embodiments, calculating a comprehensive index value according to the out-of-stock status of the product and eliminating the products to be deleted based on the comprehensive index value includes:

[0027] Determine the weight of each category of goods based on the out-of-stock status of the goods;

[0028] Calculate the comprehensive index value of each category of goods based on the weights;

[0029] Determine, based on the comprehensive index value, products in each category that are below a preset threshold;

[0030] Sort the products of each category in ascending order according to the cross ratio;

[0031] Filter out the products to be deleted based on the sorting results, and remove the products to be deleted.

[0032] In some embodiments, determining the weight of each category of goods based on the out-of-stock status of the goods includes:

[0033] Determine the initial weight of each category of goods;

[0034] Determine the out-of-stock rate of each category of goods based on the out-of-stock status of the goods;

[0035] Comparing the out-of-stock rate with the out-of-stock rate threshold corresponding to each category of goods;

[0036] The initial weights are adjusted based on the comparison.

[0037] In some embodiments, locating the positions of the remaining products using the blockchain traceability code and triggering electronic price tags to adjust prices or stepper motors to adjust shelves based on the positions include:

[0038] Based on the blockchain traceability code, the blockchain node is called to query the shelf coordinates associated with the remaining several products;

[0039] determining the locations of the remaining plurality of commodities according to the above;

[0040] Determining whether there is a commodity meeting a preset condition among the remaining commodities;

[0041] If there is a commodity that meets the preset conditions, the stepper motor is started to adjust the commodity that meets the preset conditions to the set position;

[0042] The camera on the top of the linked shelf detects the length of time the customer stays based on the positions of the remaining products;

[0043] If the stay time exceeds the preset stay time and the product is not purchased, the electronic price tag is triggered to reduce the price of the product at the corresponding position.

[0044] In addition, to achieve the above-mentioned purpose, the present invention also proposes a convenience store intelligent management device based on multimodal perception and blockchain, which includes:

[0045] The acquisition module is used to collect the weight, image and identification data of goods on the convenience store shelves in real time through weight sensors, 3D structured light vision sensors and RFID readers deployed on the shelves;

[0046] a processing module, configured to identify a product out-of-stock status and a product stacking status based on the product image, and obtain a real-time inventory quantity based on the product weight, the product out-of-stock status, and the product stacking status;

[0047] The processing module is used to classify the goods into categories, and based on the classification results, associate the product identification data with the goods to generate a blockchain traceability code;

[0048] The processing module is configured to input the real-time inventory into the LSTM model to predict sales, calculate a comprehensive index value according to the out-of-stock status of the product, and eliminate the products to be deleted based on the comprehensive index value;

[0049] The management module is used to locate the positions of the remaining commodities through the blockchain traceability code, and trigger the electronic price tags to adjust prices or the stepper motors to adjust shelves based on the positions to achieve intelligent management.

[0050] In addition, to achieve the above-mentioned purpose, the present invention also proposes a convenience store intelligent management device based on multimodal perception and blockchain, and the convenience store intelligent management device based on multimodal perception and blockchain includes: a memory, a processor, and a convenience store intelligent management program based on multimodal perception and blockchain stored on the memory and executable on the processor. The convenience store intelligent management program based on multimodal perception and blockchain is configured to implement the steps of the convenience store intelligent management method based on multimodal perception and blockchain as described above.

[0051] In addition, to achieve the above-mentioned purpose, the present invention also proposes a storage medium, on which a convenience store intelligent management program based on multimodal perception and blockchain is stored. When the convenience store intelligent management program based on multimodal perception and blockchain is executed by a processor, the steps of the convenience store intelligent management method based on multimodal perception and blockchain as described above are implemented.

[0052] The present invention uses weight sensors, 3D structured light vision sensors and RFID readers deployed on shelves to collect the weight, image and identification data of goods on convenience store shelves in real time; identifies the out-of-stock status and stacking status of goods based on the goods images, and obtains the real-time inventory based on the weight, out-of-stock status and stacking status of the goods; divides the goods into categories, and associates the goods identification data with the goods based on the classification results to generate a blockchain traceability code; inputs the real-time inventory into an LSTM model to predict sales, calculates a comprehensive index value based on the out-of-stock status of the goods, and eliminates the goods to be deleted based on the comprehensive index value; locates the positions of the remaining several goods through the blockchain traceability code, and triggers the electronic price tag to adjust the price or the stepper motor to adjust the shelf based on the position to achieve intelligent management. Through the above method, the out-of-stock status can be perceived in time to avoid replenishment delays, and at the same time, the classification of goods can be optimized and managed. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flow chart of the first embodiment of the convenience store intelligent management method based on multimodal perception and blockchain of the present invention;

[0054] Figure 2 This is a structural block diagram of the first embodiment of the convenience store intelligent management device based on multimodal perception and blockchain of the present invention.

[0055] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0056] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0057] The embodiment of the present invention provides a convenience store intelligent management method based on multimodal perception and blockchain, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of a convenience store intelligent management method based on multimodal perception and blockchain in the present invention.

[0058] In this embodiment, the convenience store intelligent management method based on multimodal perception and blockchain includes the following steps:

[0059] Step S10: The weight sensors, 3D structured light vision sensors and RFID readers deployed on the shelves are used to collect the weight, image and identification data of the goods on the convenience store shelves in real time.

[0060] In this embodiment, the executor of this embodiment is a convenience store intelligent management device based on multimodal perception and blockchain, wherein the convenience store intelligent management device based on multimodal perception and blockchain has functions such as data processing, data communication and program running. The convenience store intelligent management device based on multimodal perception and blockchain can be a computer terminal device or other network device, and of course it can also be other devices with similar functions, and this embodiment does not limit this.

[0061] It's important to note that convenience store chains, as a key component of the retail industry, rely heavily on precise category management and inventory control for operational efficiency. Traditionally, merchandise management relies primarily on manual inventory checks and empirical judgment. However, manual inventory checks are time-consuming and lack real-time detection of out-of-stock and misplaced items, leading to replenishment delays. Furthermore, product elimination relies on static rules, without incorporating dynamic out-of-stock rates, category role weighting, and other comprehensive indicators, resulting in a large number of unsold products. This low level of intelligence leads to replenishment delays and a lack of category optimization in traditional convenience store management models.

[0062] In order to solve the above technical problems, this embodiment uses weight sensors, 3D structured light vision sensors and RFID readers deployed on the shelves to collect the weight, image and identification data of goods on the convenience store shelves in real time; based on the product image, the out-of-stock status and stacking status of the goods are identified, and the real-time inventory is obtained based on the weight, out-of-stock status and stacking status of the goods; the goods are divided into categories, and the product identification data is associated with the goods based on the classification results to generate a blockchain traceability code; the real-time inventory is input into the LSTM model to predict sales, a comprehensive index value is calculated according to the out-of-stock status of the goods, and the goods to be deleted are eliminated based on the comprehensive index value; the positions of the remaining goods are located through the blockchain traceability code, and the electronic price tags are triggered to adjust the prices or the stepper motors are triggered to adjust the shelves based on the positions to achieve intelligent management. Through the above method, the out-of-stock status can be perceived in time to avoid replenishment delays, and the classification of goods can also be optimized. Specifically, it can be achieved as follows.

[0063] In a specific implementation, this embodiment first uses weight sensors, 3D structured light vision sensors, and RFID readers deployed on the shelves to collect real-time data on the weight, image, and identification of goods on the convenience store shelves. The weight sensors can be high-precision strain gauges that monitor weight changes of goods on the shelves in real time (with an accuracy of ±0.5g). The strain gauge deformation is converted into an electrical signal to calculate the total weight of the goods. The 3D structured light vision sensor can capture the three-dimensional image and spatial position of the goods. By projecting a coded light pattern, the camera captures the deformed pattern and reconstructs a 3D model, which is then used by a deep learning algorithm to compare against standard display templates. The RFID reader can be an ultra-high frequency (UHF) reader of 920–925MHz, capable of batch reading electronic tags (RFID) of goods. After activating the tag chip with an electromagnetic field, the product identification data is wirelessly transmitted.

[0064] Step S20: identifying the out-of-stock status and stacking status of the product based on the product image, and obtaining the real-time inventory based on the product weight, the out-of-stock status of the product, and the stacking status of the product.

[0065] In a specific implementation, the product images obtained above can be used to identify out-of-stock and stacked items. For example, a convolutional neural network (CNN) is used to train a product detection model (with a dataset containing over 100,000 labeled samples) to identify product bounding boxes. If the bounding box confidence level for a preset coordinate (e.g., [A3-2-4]) is less than 0.05 (no product features), the image is marked as out-of-stock. If the height change is greater than the thickness of a single item (e.g., stacked beverage cans), the image is considered stacked. If the normal vector tilt angle is greater than 15°, the image is marked as tilted.

[0066] Furthermore, based on the commodity weight, the commodity out-of-stock status and the commodity stacking status, obtaining the real-time inventory quantity specifically includes locating the out-of-stock area based on the commodity out-of-stock status; determining the mass of missing commodities in the located out-of-stock area based on the commodity weight; determining the mass of effective commodities in the non-out-of-stock area based on the mass of missing commodities; determining the effective commodity density in the non-out-of-stock area based on the commodity stacking status; and determining the real-time inventory quantity based on the effective commodity mass and the effective commodity density.

[0067] It should be noted that the goods in the out-of-stock area have their standard quality. The quality of the missing goods can be determined by identifying the out-of-stock situation and combining it with the standard quality of the goods in the out-of-stock area. The effective quality of the goods is the difference between the total weight of the goods and the quality of the missing goods, for example, Q1 = Q z -∑Q L , Q1 represents the effective product quality, Q S Indicates the total weight of the product, Q LIndicates the quality of missing goods. When determining the effective commodity density of the non-out-of-stock area, it is necessary to first call the commodity density library, through which the reference commodity density of the non-out-of-stock area can be obtained, that is, the standard density. Since stacking will cause or cause density changes, for the goods blocked by stacking in the non-out-of-stock area, this embodiment uses a preset density compensation coefficient to adjust the reference commodity density of the goods blocked by stacking, for example, dq=ds×β, where dq represents the standard density, ds represents the standard density, β represents the compensation coefficient, and the compensation coefficient can be 0.92~1.05. The effective commodity density of goods without stacking obstructions is the reference commodity density determined above. Finally, the real-time inventory can be calculated based on the effective commodity quality and the effective commodity density, S=Q1 / (dq×Qs), and Qs is the standard quality of a single commodity.

[0068] Step S30: Classify the goods into categories, and associate the product identification data with the goods based on the classification results to generate a blockchain traceability code.

[0069] In this embodiment, products are categorized according to four roles: targeted, conventional, convenience, and seasonal. Based on these categorizations, the weights of each category are extracted, and the minimum stock keeping unit (SKU) and production batch number are extracted from the product identification data. Target categories include high-frequency, essential goods, conventional categories include daily necessities, convenience categories include impulse buys, and seasonal categories include seasonal goods. Each of the four categories has a corresponding initial weight: for example, the weight of targeted categories is 0.33, the weight of conventional categories is 0.23, the weight of convenience categories is 0.17, and the weight of seasonal categories is 0.27. The minimum stock keeping unit (SKU) is derived from these two values ​​to create a unique hash value: for example, SKU + production batch number = "A12345-BATCH20250603." Finally, the hash value is associated with the weights of each category to generate a blockchain traceability code.

[0070] Step S40: Input the real-time inventory into the LSTM model to predict sales, calculate a comprehensive index value according to the out-of-stock status of the product, and eliminate the products to be deleted based on the comprehensive index value.

[0071] It should be understood that the convenience store management in this embodiment includes price adjustment and shelf adjustment. Before performing these two operations, some products need to be removed, that is, products are prepared for deletion.

[0072] In a specific implementation, the final weights of each category of goods can be obtained by combining the initial weights of the aforementioned categories of goods and the out-of-stock status of the goods. Specifically, the initial weights of the weights of each category of goods are determined; the out-of-stock rate of each category of goods is determined based on the out-of-stock status of the goods; the out-of-stock rate is compared with the out-of-stock rate threshold corresponding to each category of goods; and the initial weights are adjusted based on the comparison.

[0073] It should be noted that the out-of-stock count count can be calculated based on the product's out-of-stock status. The out-of-stock rate can be calculated by comparing the ratio of the out-of-stock count count to the total inventory count. Different out-of-stock rate thresholds for different product categories can be used to adjust weights. For example, if the initial weight of a targeted category is 0.33, when the out-of-stock rate exceeds 5%, the weight is multiplied by 0.8. Since 5% is the out-of-stock rate threshold for the targeted category, the weight becomes 0.264. For another example, if the initial weight of a conventional category is 0.23, when the out-of-stock rate exceeds 10%, the weight is multiplied by 0.7.

[0074] Furthermore, the calculation formula of the comprehensive index value is f ij (X,W,Y)=αX+W*0.16+Y*0.51, where X represents sales, W represents sales volume, Y represents gross profit, and α represents the weight of each category of goods. 0.16 and 0.51 in the calculation formula can be set according to actual needs, and there is no restriction on this in this embodiment. After calculating the comprehensive index value, the goods in each category that are lower than the preset threshold are then determined; the goods in each category are sorted in ascending order according to the cross ratio, and the goods to be deleted are screened out based on the sorting results, and the goods to be deleted are eliminated. Specifically, the cross ratio is turnover rate × gross profit margin. After sorting, the 5% with the lowest cross ratio are selected as goods to be deleted.

[0075] Step S50: Locate the positions of the remaining products through the blockchain traceability code, and trigger the electronic price tags to adjust prices or the stepper motors to adjust shelves based on the positions to achieve intelligent management.

[0076] In a specific implementation, based on the blockchain traceability code, the blockchain node is called to query the shelf coordinates associated with the remaining several commodities, and the positions of the remaining several commodities are determined according to the blockchain traceability code (format: CCM-{hash value}-{category role code}), and the blockchain node is called to query the associated shelf coordinates (XYZ three-dimensional coordinates). The traceability code CCM-4f7c9a21b-T01 corresponds to the coordinates [A3-2-4] (shelf No. 3, area A, 2nd layer, 4th column).

[0077] Furthermore, in this embodiment, the products that meet the preset conditions are hot-selling products. The preset conditions can be products with a cross ratio > 2.0. Then the stepper motor is started to lower the height of the shelf layer, and the layer is raised 20cm to the prime display position. The camera on the top of the linked shelf detects the length of time the customer stays based on the position of the remaining products; if the length of time exceeds the preset length of time and the product is not purchased, the electronic price tag is triggered to reduce the price of the product at the corresponding position. For example, the length of time the customer stays in the coordinate area is detected in real time; if the stay is > 15 seconds and no purchase is made, the electronic price tag is triggered to reduce the price for the second time, reducing the price by 10%. The above preset time and price reduction range can be adjusted according to actual conditions, and this is not limited in this embodiment.

[0078] This embodiment uses weight sensors, 3D structured light vision sensors and RFID readers deployed on the shelves to collect the weight, image and identification data of goods on the convenience store shelves in real time; identifies the out-of-stock status and stacking status of goods based on the product images, and obtains the real-time inventory based on the product weight, out-of-stock status and stacking status; divides the goods into categories, and associates the product identification data with the goods based on the classification results to generate a blockchain traceability code; inputs the real-time inventory into the LSTM model to predict sales, calculates a comprehensive index value based on the out-of-stock status of the goods, and eliminates the goods to be deleted based on the comprehensive index value; locates the positions of the remaining goods through the blockchain traceability code, and triggers the electronic price tag price adjustment or the stepper motor to adjust the shelf based on the position to achieve intelligent management. In this way, the out-of-stock status can be perceived in time to avoid replenishment delays, and the classification of goods can also be optimized.

[0079] In addition, an embodiment of the present invention also proposes a storage medium, on which a convenience store intelligent management program based on multimodal perception and blockchain is stored. When the convenience store intelligent management program based on multimodal perception and blockchain is executed by a processor, the steps of the convenience store intelligent management method based on multimodal perception and blockchain as described above are implemented.

[0080] Reference Figure 2 , Figure 2 This is a structural block diagram of the first embodiment of the convenience store intelligent management device based on multimodal perception and blockchain of the present invention.

[0081] like Figure 2 As shown, the convenience store intelligent management device based on multimodal perception and blockchain proposed in an embodiment of the present invention includes:

[0082] The acquisition module 10 is used to collect the weight, image and identification data of goods on the convenience store shelves in real time through the weight sensors, 3D structured light vision sensors and RFID readers deployed on the shelves;

[0083] a processing module 20 for identifying a product out-of-stock status and a product stacking status based on the product image, and obtaining a real-time inventory quantity based on the product weight, the product out-of-stock status, and the product stacking status;

[0084] The processing module 20 is used to classify the goods into categories, and based on the classification results, associate the product identification data with the goods to generate a blockchain traceability code;

[0085] The processing module 20 is configured to input the real-time inventory into the LSTM model to predict sales, calculate a comprehensive index value according to the out-of-stock status of the product, and eliminate the products to be deleted based on the comprehensive index value;

[0086] The management module 30 is used to locate the positions of the remaining commodities through the blockchain traceability code, and trigger the electronic price tags to adjust prices or the stepper motors to adjust shelves based on the positions to achieve intelligent management.

[0087] This embodiment uses weight sensors, 3D structured light vision sensors and RFID readers deployed on the shelves to collect the weight, image and identification data of goods on the convenience store shelves in real time; identifies the out-of-stock status and stacking status of goods based on the product images, and obtains the real-time inventory based on the product weight, out-of-stock status and stacking status; divides the goods into categories, and associates the product identification data with the goods based on the classification results to generate a blockchain traceability code; inputs the real-time inventory into the LSTM model to predict sales, calculates a comprehensive index value based on the out-of-stock status of the goods, and eliminates the goods to be deleted based on the comprehensive index value; locates the positions of the remaining goods through the blockchain traceability code, and triggers the electronic price tag price adjustment or the stepper motor to adjust the shelf based on the position to achieve intelligent management. In this way, the out-of-stock status can be perceived in time to avoid replenishment delays, and the classification of goods can also be optimized.

[0088] In some embodiments, the processing module 20 is configured to locate an out-of-stock area based on the out-of-stock status of the product;

[0089] Determining the mass of the missing goods in the located out-of-stock area based on the weight of the goods;

[0090] Determine the effective product quality in the non-out-of-stock area based on the missing product quality;

[0091] Determining an effective commodity density in a non-out-of-stock area based on the commodity stacking status;

[0092] The real-time inventory quantity is determined according to the effective commodity quality and the effective commodity density.

[0093] In some embodiments, the processing module 20 is configured to call a commodity density library to determine a reference commodity density in a non-out-of-stock area;

[0094] Determining, based on the stacking status of the goods, goods that are blocked by stacking and goods that are not blocked by stacking in the non-out-of-stock area;

[0095] The reference commodity density of the commodities not blocked by stacking is used as the effective commodity density, and the reference commodity density of the commodities blocked by stacking is adjusted by using a preset density compensation coefficient to obtain the effective commodity density.

[0096] In some embodiments, the processing module 20 is configured to classify the products into four categories, wherein the four categories include target categories, regular categories, convenience categories, and seasonal categories.

[0097] Extracting the weight of each category of goods based on the classification results, and extracting the minimum stock unit and production batch number of the goods from the said product identification data;

[0098] Generate a unique hash value based on the minimum stock keeping unit of the product and the production batch number using the SHA-256 algorithm;

[0099] The hash value is associated with the weight of each category of goods, and is written into the blockchain through the PBFT consensus mechanism to obtain a blockchain traceability code.

[0100] In some embodiments, the processing module 20 is configured to determine the weight of each category of goods according to the out-of-stock status of the goods;

[0101] Calculate the comprehensive index value of each category of goods based on the weights;

[0102] Determine, based on the comprehensive index value, products in each category that are below a preset threshold;

[0103] Sort the products of each category in ascending order according to the cross ratio;

[0104] Filter out the products to be deleted based on the sorting results, and remove the products to be deleted.

[0105] In some embodiments, the processing module 20 is used to determine the initial weight of each category of goods;

[0106] Determine the out-of-stock rate of each category of goods based on the out-of-stock status of the goods;

[0107] Comparing the out-of-stock rate with the out-of-stock rate threshold corresponding to each category of goods;

[0108] The initial weights are adjusted based on the comparison.

[0109] In some embodiments, the management module 30 is configured to call a blockchain node based on the blockchain traceability code to query the shelf coordinates associated with the remaining plurality of commodities;

[0110] determining the locations of the remaining plurality of commodities according to the above;

[0111] Determining whether there is a commodity meeting a preset condition among the remaining commodities;

[0112] If there is a commodity that meets the preset conditions, the stepper motor is started to adjust the commodity that meets the preset conditions to the set position;

[0113] The camera on the top of the linked shelf detects the length of time the customer stays based on the positions of the remaining products;

[0114] If the stay time exceeds the preset stay time and the product is not purchased, the electronic price tag is triggered to reduce the price of the product at the corresponding position.

[0115] An embodiment of the present application also provides a convenience store intelligent management device based on multimodal perception and blockchain, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus, the memory is used to store the convenience store intelligent management program based on multimodal perception and blockchain; the processor is used to implement the above-mentioned convenience store intelligent management method based on multimodal perception and blockchain when executing the program stored in the memory.

[0116] The communication bus mentioned in the aforementioned multimodal sensing and blockchain-based convenience store smart management device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into an address bus, a data bus, and a control bus.

[0117] The communication interface is used for communication between the above-mentioned convenience store intelligent management device based on multimodal perception and blockchain and other devices.

[0118] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage. Alternatively, the memory may be at least one storage device located remote from the processor.

[0119] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0120] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0121] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0122] Each embodiment in this specification is described in a related manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiment is generally similar to the method embodiment, so the description is relatively simple. For related parts, refer to the description of the method embodiment.

[0123] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

[0124] It should be understood that the above is only an example and does not constitute any limitation to the technical solution of the present invention. In specific applications, those skilled in the art can make settings as needed, and the present invention does not impose any limitation on this.

[0125] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of it according to actual needs to achieve the purpose of the embodiment scheme, and no limitation is made here.

[0126] In addition, for technical details not fully described in this embodiment, please refer to the convenience store intelligent management method based on multimodal perception and blockchain provided in any embodiment of the present invention, and will not be repeated here.

[0127] In addition, it should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0128] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0129] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, or of course by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as a read-only memory (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0130] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

[0131] It is understandable that the system provided by the embodiment of the present invention corresponds to the method provided by the embodiment of the present invention, and the explanation, examples and beneficial effects of the relevant contents can refer to the corresponding parts of the above method.

Claims

1. A convenience store intelligent management method based on multimodal perception and blockchain, characterized by: The convenience store intelligent management method based on multimodal perception and blockchain includes: The weight sensors, 3D structured light vision sensors and RFID readers deployed on the shelves collect the weight, image and identification data of the goods on the convenience store shelves in real time; identifying a product out-of-stock status and a product stacking status based on the product image, and obtaining a real-time inventory quantity based on the product weight, the product out-of-stock status, and the product stacking status; Classify the goods into categories, and based on the classification results, associate the product identification data with the goods to generate a blockchain traceability code; Input the real-time inventory into the LSTM model to predict sales, calculate a comprehensive index value based on the out-of-stock status of the product, and eliminate the products to be deleted based on the comprehensive index value; The locations of the remaining commodities are located through the blockchain traceability code, and electronic price tags are triggered to adjust prices or stepper motors are triggered to adjust shelves based on the locations to achieve intelligent management.

2. The convenience store intelligent management method based on multimodal perception and blockchain according to claim 1 is characterized in that: The obtaining of the real-time inventory quantity based on the weight of the commodity, the out-of-stock status of the commodity, and the stacking status of the commodity includes: Locating an out-of-stock area based on the out-of-stock status of the product; Determining the mass of the missing goods in the located out-of-stock area based on the weight of the goods; Determine the effective product quality in the non-out-of-stock area based on the missing product quality; Determining an effective commodity density in a non-out-of-stock area based on the commodity stacking status; The real-time inventory quantity is determined according to the effective commodity quality and the effective commodity density.

3. The convenience store intelligent management method based on multimodal perception and blockchain according to claim 2 is characterized in that: The determining of the effective commodity density in the non-out-of-stock area based on the commodity stacking state includes: Call the commodity density database to determine the reference commodity density in the non-out-of-stock area; Determining, based on the stacking status of the goods, goods that are blocked by stacking and goods that are not blocked by stacking in the non-out-of-stock area; The reference commodity density of the commodities not blocked by stacking is used as the effective commodity density, and the reference commodity density of the commodities blocked by stacking is adjusted by using a preset density compensation coefficient to obtain the effective commodity density.

4. The convenience store intelligent management method based on multimodal perception and blockchain according to claim 1 is characterized in that: The classification of commodities into categories, and associating the commodity identification data with the commodities based on the classification results to generate a blockchain traceability code, include: Classify the products into four categories, including target categories, regular categories, convenience categories, and seasonal categories; Extracting the weight of each category of goods based on the classification results, and extracting the minimum stock unit and production batch number of the goods from the said product identification data; Generate a unique hash value based on the minimum stock keeping unit of the product and the production batch number using the SHA-256 algorithm; The hash value is associated with the weight of each category of goods, and is written into the blockchain through the PBFT consensus mechanism to obtain a blockchain traceability code.

5. The convenience store intelligent management method based on multimodal perception and blockchain according to claim 1 is characterized in that: The calculating of a comprehensive index value according to the out-of-stock status of the product and eliminating the products to be deleted based on the comprehensive index value includes: Determine the weight of each category of goods based on the out-of-stock status of the goods; Calculate the comprehensive index value of each category of goods based on the weights; Determine, based on the comprehensive index value, products in each category that are below a preset threshold; Sort the products of each category in ascending order according to the cross ratio; Filter out the products to be deleted based on the sorting results, and remove the products to be deleted.

6. The convenience store intelligent management method based on multimodal perception and blockchain according to claim 5 is characterized in that: Determining the weight of each category of goods according to the out-of-stock status of the goods includes: Determine the initial weight of each category of goods; Determine the out-of-stock rate of each category of goods based on the out-of-stock status of the goods; Comparing the out-of-stock rate with the out-of-stock rate threshold corresponding to each category of goods; The initial weights are adjusted based on the comparison.

7. The convenience store intelligent management method based on multimodal perception and blockchain according to claim 1 is characterized in that: The method of locating the positions of the remaining commodities by using the blockchain traceability code and triggering electronic price tags to adjust prices or stepper motors to adjust shelves based on the positions includes: Based on the blockchain traceability code, the blockchain node is called to query the shelf coordinates associated with the remaining several products; determining the locations of the remaining plurality of commodities according to the above; Determining whether there is a commodity meeting a preset condition among the remaining commodities; If there is a commodity that meets the preset conditions, the stepper motor is started to adjust the commodity that meets the preset conditions to the set position; The camera on the top of the linked shelf detects the length of time the customer stays based on the positions of the remaining products; If the stay time exceeds the preset stay time and the product is not purchased, the electronic price tag is triggered to reduce the price of the product at the corresponding position.

8. A convenience store intelligent management device based on multimodal perception and blockchain, characterized in that: The convenience store intelligent management device based on multimodal perception and blockchain includes: The acquisition module is used to collect the weight, image and identification data of goods on the convenience store shelves in real time through weight sensors, 3D structured light vision sensors and RFID readers deployed on the shelves; a processing module, configured to identify a product out-of-stock status and a product stacking status based on the product image, and obtain a real-time inventory quantity based on the product weight, the product out-of-stock status, and the product stacking status; The processing module is used to classify the goods into categories, and based on the classification results, associate the product identification data with the goods to generate a blockchain traceability code; The processing module is configured to input the real-time inventory into the LSTM model to predict sales, calculate a comprehensive index value according to the out-of-stock status of the product, and eliminate the products to be deleted based on the comprehensive index value; The management module is used to locate the positions of the remaining commodities through the blockchain traceability code, and trigger the electronic price tags to adjust prices or the stepper motors to adjust shelves based on the positions to achieve intelligent management.

9. A convenience store intelligent management device based on multimodal perception and blockchain, characterized by: The convenience store intelligent management device based on multimodal perception and blockchain includes: a memory, a processor, and a convenience store intelligent management program based on multimodal perception and blockchain stored on the memory and executable on the processor. The convenience store intelligent management program based on multimodal perception and blockchain is configured to implement the steps of the convenience store intelligent management method based on multimodal perception and blockchain as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a convenience store intelligent management program based on multimodal perception and blockchain. When the convenience store intelligent management program based on multimodal perception and blockchain is executed by the processor, the steps of the convenience store intelligent management method based on multimodal perception and blockchain as described in any one of claims 1 to 7 are implemented.

Images