Self-adaptive rapid development self-service cashier method and system, medium and program product

Abstract

The invention provides a self-adaptive rapid development self-service cashier method and system, a medium and a program product, and relates to the technical field of intelligent self-service cashier, and the method comprises the steps: firstly obtaining the information of a commodity photo and a retail scene photo, determining a current sales scene through a deep learning training model, obtaining the information of hardware equipment to determine a payment mode, and carrying out the self-service cashier. A commodity pricing mode is determined in combination with a sales scene and a commodity photo, a self-service cashier system infrastructure is determined according to the sales scene, a payment mode and the pricing mode, and a commodity display module, a payment processing module and a pricing calculation module are included. And finally, according to a user instruction of a cashier display end, modifying an interface style, color matching and button layout by using the determined system infrastructure. Therefore, the self-service cashier can automatically adapt to different retail scenes, and the flexibility of the self-service cashier system is improved.

Description

Technical Field

[0001] This application relates to the technical field of intelligent self-checkout, and particularly to an adaptive rapid development self-checkout method, system, medium, and program product. Background Art

[0002] With the rapid development of the retail industry, self-checkout systems have become an important means to improve shopping efficiency and reduce labor costs. Especially in retail scenarios such as large supermarkets and convenience stores, the application of self-checkout systems is becoming increasingly popular.

[0003] Currently, common self-checkout systems adopt a fixed system architecture and interface design. When the system is deployed, developers need to perform customized development according to specific application scenarios. This development mode usually uses preset templates, and developers manually configure the product display method, payment function, and pricing rules, and then adjust the interface layout for different retail scenarios.

[0004] However, due to the diversity and dynamic change characteristics of retail scenarios, this fixed-template development method often requires frequent manual intervention and adjustment in actual applications. When the retail scenario changes, the system cannot adapt to the new scenario requirements in a timely manner, resulting in a mismatch between the system functions and the actual application scenarios, affecting the user experience and checkout efficiency. Summary of the Invention

[0005] This application provides an adaptive rapid development self-checkout method, system, medium, and program product, which is used to solve the problem that traditional self-checkout systems are difficult to quickly adapt to business changes and usage scenarios when facing complex and diverse retail scenarios.

[0006] In a first aspect, this application provides an adaptive rapid development self-checkout method, which is applied to a self-checkout system. The method includes: obtaining multiple product photo information and retail scenario photo information; determining the current sales scenario information according to the multiple product photo information and the retail scenario photo information, in combination with a sales scenario recognition model, which is pre-trained by deep learning according to multiple labeled sales scenario photos; after obtaining multiple hardware device information, determining the payment method information of the product in combination with the multiple hardware device information, and the payment method information at least includes face recognition payment method; determining the product pricing method in combination with the current sales scenario information and the product photo information; determining the basic architecture of the self-checkout system according to the current sales scenario information, the payment method information, and the product pricing method, and the basic architecture of the self-checkout system at least includes a product display module, a payment processing module, and a pricing calculation module; after receiving a user instruction sent by the checkout display terminal, modifying the interface style, color matching, and operation button layout of the checkout display terminal in combination with the basic architecture of the self-checkout system.

[0007] By adopting the above technical solution, the retail scene photos are deeply analyzed through the sales scene recognition model to accurately identify the current sales scene features. The system automatically matches the most suitable payment method and commodity pricing method according to the scene features and hardware device information, and then constructs the basic architecture of the self-service cash register system that meets the actual needs. This basic architecture can flexibly adjust the interface style and layout according to user instructions, enabling the system to have scene adaptability. This adaptive development method significantly improves the system development efficiency, reduces the development cost, and at the same time ensures the practicability and usability of the system.

[0008] Combined with some embodiments of the first aspect, in some embodiments, after the step of determining the basic architecture of the self-service cash register system according to the current sales scene information, the payment method information, and the commodity pricing method, it further includes: determining business rule information according to the sales scene information and the commodity pricing method, where the business rule information at least includes promotion activity rules, membership point rules, and inventory management rules; after receiving the user's instruction to modify the rules, updating the business rule information.

[0009] By adopting the above technical solution, the system can automatically generate business rule information including promotion activities, membership points, etc. according to the sales scene information and the commodity pricing method. These rules can be updated and adjusted in real time according to user needs, enabling the system to have the flexibility of business logic while maintaining the stability of the basic architecture. This dynamic rule management mechanism improves the operation efficiency of the system, enhances the business adaptability, and enables the self-service cash register system to quickly respond to market changes and user needs.

[0010] Combined with some embodiments of the first aspect, in some embodiments, after the step of determining the business rule information according to the sales scene information and the commodity pricing method, it further includes: using natural language processing technology to collect and analyze the information set sent by the cash register display terminal, where the information set includes evaluation information, consultation information input by the user, and voice interaction information in the retail scene; combining the current sales scene information and the commodity pricing method, performing sentiment analysis and intention recognition on the information set; if it is recognized that the customer has a set demand for the target commodity, then feedback according to the set demand to the merchant management terminal; after receiving the adjustment information sent by the merchant management terminal, adjusting the business rule information in real time according to the adjustment information.

[0011] By adopting the above technical solution, the user feedback is deeply analyzed by means of natural language processing technology to accurately identify the user's intention and sentiment tendency. The system can timely capture the user's demand for specific commodities and quickly feedback this information to the merchant. The merchant can adjust the business rules accordingly to achieve a precise response to the user's needs. This intelligent adjustment mechanism based on user feedback greatly improves the user satisfaction of the system and at the same time provides a more accurate basis for the merchant's business decision-making.

[0012] In some embodiments in combination with some embodiments of the first aspect, before the step of determining the self-service cash register system infrastructure according to the current sales scenario information, the payment method information, and the commodity pricing method, it further includes: extracting commodity features from the commodity photo information based on a deep learning model to obtain a set of commodity feature information, where the set of commodity feature information at least includes commodity appearance features, packaging features, and brand logo features; classifying the commodities by combining the set of commodity feature information and preset commodity classification rules to generate commodity classification labels; generating a commodity display plan according to the commodity classification labels and the current sales scenario information, where the commodity display plan includes commodity placement positions, combined commodity displays, and key commodity recommendation areas; dynamically adjusting the display logic of the commodity display module in the self-service cash register system infrastructure based on the commodity display plan.

[0013] By adopting the above technical solution, the deep learning model comprehensively extracts commodity features and generates an optimal commodity display plan in combination with scenario information. This intelligent commodity management method can automatically optimize the display logic and layout of commodities according to commodity characteristics and sales scenarios. The system not only realizes intelligent classification and accurate recommendation of commodities, but also can dynamically adjust the display strategy according to the actual situation, significantly improving the commodity display effect and user shopping experience.

[0014] In some embodiments in combination with some embodiments of the first aspect, before the step of determining the self-service cash register system infrastructure according to the current sales scenario information, the payment method information, and the commodity pricing method, it further includes: constructing a commodity price prediction model, where the commodity price prediction model is trained based on a deep learning algorithm by analyzing historical transaction data and seasonal factor data; collecting market price data in real time, inputting the market price data into the commodity price prediction model to obtain a commodity price prediction result; automatically adjusting the commodity pricing strategy in the pricing calculation module according to the commodity price prediction result and in combination with the current sales scenario information; By adopting the above technical solution, the commodity price prediction model provides a scientific basis for price adjustment by analyzing historical data and real-time market data. The model can automatically adjust the pricing strategy according to the prediction result to ensure the market competitiveness of the price. This intelligent pricing mechanism based on deep learning not only improves the accuracy and timeliness of pricing, but also can effectively respond to market changes and optimize the efficiency of commodity price management.

[0015] In some embodiments in combination with some embodiments of the first aspect, after the step of modifying the interface style, color scheme, and operation button layout of the cashier display terminal in combination with the basic infrastructure of the self-checkout system after receiving a user instruction from the cashier display terminal, the method further includes: analyzing the operation behavior data of the user by using a machine learning algorithm, where the operation behavior data includes click hotspots, residence duration, and operation paths; constructing a user experience evaluation model based on the operation behavior data, where the user experience evaluation model is used to evaluate the usability metrics of interface interaction in real time; optimizing the interface layout and interaction process according to the usability metrics to generate a personalized interface adaptation solution; and when a set user group is detected through a camera, controlling the cashier display terminal to automatically switch to the corresponding interface adaptation solution.

[0016] By adopting the above technical solution, the machine learning algorithm analyzes the user operation behavior data, evaluates the usability of interface interaction in real time, the system automatically optimizes the interface layout and interaction process according to the evaluation result, and can automatically switch the interface solution for different user groups. This intelligent interface adaptation mechanism significantly improves the interaction experience of the system, makes the self-checkout operation more convenient and intuitive, and greatly improves user satisfaction.

[0017] In some embodiments in combination with some embodiments of the first aspect, in the step of determining the current sales scenario information by combining the sales scenario recognition model according to multiple pieces of the commodity photo information and the retail scenario photo information, it specifically includes: while obtaining multiple pieces of the commodity photo information and the retail scenario photo information through a camera, obtaining ambient illuminance data and passenger flow data through a sensor; establishing a scene feature vector by combining the ambient illuminance data and the passenger flow data; inputting the scene feature vector into the sales scenario recognition model to output a scene classification result; and determining the current sales scenario information according to the scene classification result.

[0018] By adopting the above technical solution, the system collects environmental data in real time through a camera and a sensor, constructs an accurate scene feature vector, and inputs these feature vectors into the sales scenario recognition model, which can more accurately identify and classify the current scene. This multi-dimensional scene recognition method improves the accuracy of scene judgment, provides a reliable decision-making basis for subsequent system configuration and function adjustment, and enables the entire self-checkout system to more intelligently adapt to different scene requirements.

[0019] In a second aspect, the present application provides a self-checkout system, where the self-checkout system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the self-checkout system to execute the methods described in the first aspect and any possible implementation manner in the first aspect.

[0020] In a third aspect, the present application provides a computer-readable storage medium including instructions that, when running on a self-checkout system, cause the self-checkout system to execute the method described in the first aspect and any possible implementation manner of the first aspect.

[0021] In a fourth aspect, the present application provides a computer program product that, when running on a self-checkout system, causes the self-checkout system to execute the method described in the first aspect and any possible implementation manner of the first aspect.

[0022] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By adopting the above technical solution, since a sales scenario recognition model based on deep learning is used for scenario feature analysis and a technical means of dynamically constructing a system architecture in combination with hardware device information is adopted, the technical problem that the self-checkout system in the prior art cannot flexibly adjust the system structure according to different sales scenarios is effectively solved. Furthermore, the scenario adaptive development of the self-checkout system is realized, the system development efficiency is significantly improved, the development cost is reduced, and at the same time, the practicability and usability of the system in different scenarios are ensured.

[0023] 2. By adopting the above technical solution, since a user feedback analysis technology based on natural language processing is used and the sentiment analysis and intention recognition results are fed back to the merchant management end in real time for business rule adjustment, the technical problem that the self-checkout system in the prior art cannot accurately identify and quickly respond to changes in user needs is effectively solved. Furthermore, the intelligent business rule adjustment based on user feedback is realized, the user satisfaction of the system is greatly improved, and it is ensured that the merchant can adjust the business strategy in a timely and accurate manner.

[0024] 3. By adopting the above technical solution, since a commodity price prediction model based on a deep learning algorithm is used and a technical means of automatically adjusting the price strategy in combination with real-time market data is adopted, the technical problem that the price management of the self-checkout system in the prior art lags behind and cannot quickly respond to market changes is effectively solved. Furthermore, the intelligent commodity pricing management is realized, the accuracy and timeliness of price adjustment are significantly improved, and the market competitiveness of commodity prices is ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 is a flowchart of an adaptive fast development self-checkout method in an embodiment of the present application; Figure 2 is another flowchart of an adaptive fast development self-checkout method in an embodiment of the present application; Figure 3It is a schematic structural diagram of an entity device in the self-checkout system according to an embodiment of the present application. Detailed implementation manners

[0026] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. As used in the specification and appended claims of the present application, the singular forms "a", "an", "the", "above-mentioned", "said", and "this" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more of the listed items.

[0027] Hereinafter, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as implying or suggesting relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise specified, the meaning of "a plurality" is two or more.

[0028] For ease of understanding, the method provided in this embodiment is described in a process below. Please refer to Figure 1 , which is a schematic flowchart of an adaptive rapid development self-checkout method according to an embodiment of the present application.

[0029] S101. Determine the current sales scenario information according to a plurality of pieces of the commodity photo information and the retail scenario photo information, in combination with a sales scenario recognition model, where the sales scenario recognition model is pre-trained by deep learning according to a plurality of labeled sales scenario photos; The self-checkout system first uses high-definition cameras installed at different positions in the store to take all-round photos of the commodity display area, the customer shopping area, and the entire retail scenario, to obtain a plurality of pieces of commodity photo information and retail scenario photo information, and these photo information are transmitted in real time to the image preprocessing module of the system through a high-speed network. In the image preprocessing module, an advanced image enhancement algorithm is used to process the photos. This algorithm can improve the resolution of the photos without losing image details, making the features of the commodities and the details in the scenario more clearly distinguishable.

[0030] The pre - processed photo information is input into the sales scenario recognition model. This model is pre - trained in advance based on a large number of labeled sales scenario photos using a deep learning framework. During the training process, transfer learning technology is adopted. Based on the model pre - trained on a large - scale image dataset, it is fine - tuned in combination with retail scenario photo data. This can make full use of the general image features learned by the pre - trained model, accelerate the convergence speed of the model, and improve the recognition accuracy. To further enhance the performance of the model, an attention mechanism is introduced. The attention mechanism enables the model to pay more attention to the key areas related to sales scenario recognition, such as the way of commodity display, customer density, store decoration style, etc., while ignoring irrelevant information, thereby improving the accuracy and efficiency of recognition. For example, when recognizing a promotion scenario, the model can focus on the commodity area with promotion signs and the crowd gathering area through the attention mechanism, and accurately determine that the current scenario is a promotion scenario.

[0031] S102. After obtaining the information of multiple hardware devices, determine the payment method information of the commodity in combination with the information of multiple such hardware devices. The payment method information includes at least the face - brushing payment method. The self - service cash register system obtains the information of multiple hardware devices connected to the system in real - time through the hardware device management module. These hardware devices include, but are not limited to, cameras, barcode scanners, face - brushing payment devices, bank card readers, mobile payment terminals, etc. The hardware device management module communicates with each hardware device through the device driver to obtain information such as the model, status, and function of the device. After obtaining the hardware device information, the system first detects the device status. For the camera, the system will check whether the image acquisition clarity and frame rate are normal; for the barcode scanner, it will detect the accuracy and response speed of barcode scanning; for the face - brushing payment device, it will verify the accuracy and recognition speed of face recognition, etc. If a certain hardware device is found to have a fault or abnormal performance, the system will automatically switch to the standby device or adjust the priority of the payment method to ensure the smooth progress of the payment process.

[0032] Combining the function and status information of the hardware devices, the system determines the payment method information of the commodity. If the system detects that the face - brushing payment device is running normally and its face recognition accuracy reaches the set standard, then the face - brushing payment method is taken as one of the preferred payment methods. At the same time, if the barcode scanner can correctly recognize various barcodes, including one - dimensional codes and two - dimensional codes, then barcode scanning payment (such as WeChat payment, Alipay payment, etc.) is also determined as an available payment method. For the bank card reader, if it can correctly read the bank card information and communicate securely with the bank system, bank card payment will also be included in the payment method list.

[0033] To improve the security and convenience of payments, the system uses a variety of encryption technologies to protect the data during the payment process. During face recognition payment, a live detection technology is used, such as live detection based on an infrared camera and deep learning algorithms, to prevent spoofing attacks with photos, videos, etc. At the same time, face data is encrypted for storage and transmission, and advanced encryption algorithms (such as the AES encryption algorithm) are used to ensure the security of face data. During QR code payment and bank card payment, payment information (such as payment amount, bank card number, payment password, etc.) is encrypted to prevent information leakage.

[0034] In addition, the system also has a function of intelligent payment method recommendation. Based on the user's historical payment records and current sales scenario information, the system will automatically recommend the most suitable payment method for the user. For example, if a user often uses WeChat Pay and the current sales scenario is shopping in a convenience store, the system will give priority to recommending WeChat Pay and prominently display the WeChat Pay icon and relevant prompt information on the cash register display terminal. During the payment process, the system will also interact with third-party payment platforms in real time to obtain the latest policies and preferential information of the payment platforms. For example, if a certain payment platform is having a full reduction activity, the system will prompt the user on the cash register display terminal to choose this payment method to enjoy the discount, improving the user's payment experience and satisfaction.

[0035] S103. Determine the commodity pricing method in combination with the current sales scenario information and the commodity photo information; After the self-service cash register system obtains the current sales scenario information and the commodity photo information, it first uses image recognition technology to process the commodity photo. During the image recognition process, advanced object detection algorithms are used to quickly and accurately identify information such as the type, brand, and specification of the commodity. Combining the identified commodity information and the current sales scenario information, the system queries the corresponding pricing rules from the commodity database. The commodity database stores detailed information of a large number of commodities, including the basic information of the commodities, the pricing methods under different sales scenarios, promotional activity information, etc. For example, in a normal sales scenario, a certain brand of beverage is priced by unit price; while in a promotional scenario, a buy-two-get-one-free combined pricing method may be launched. For some special commodities, such as fresh food commodities, in addition to using image recognition technology to obtain the type information of the commodities, the system will also combine devices such as electronic scales to obtain the weight information of the commodities and adopt a pricing method based on weight. During this process, the system will verify and process the data transmitted by the electronic scale to ensure the accuracy of the weight data.

[0036] To handle complex promotional activities, the system adopts rule engine technology. The rule engine can automatically match corresponding promotional rules based on sales scenario information and product information, and calculate the final price of the product. For example, if the current sales scenario is a holiday promotion and the product meets the full reduction activity rules, the rule engine will automatically calculate the price after the full reduction. At the same time, the rule engine also supports nested rules and combined rules, and can handle the situation where multiple promotional activities are superimposed. During the process of determining the product pricing method, the system will also consider member information. If the customer is a member, the system will adjust the product price according to the member level and member point rules.

[0037] S104. Determine the self-checkout system infrastructure according to the current sales scenario information, the payment method information, and the product pricing method. The self-checkout system infrastructure at least includes a product display module, a payment processing module, and a pricing calculation module. After obtaining the current sales scenario information, payment method information, and product pricing method, the self-checkout system starts to build the self-checkout system infrastructure. This process is dynamically adjusted based on the pre-set architecture template library of the system and the actual obtained information.

[0038] The system first analyzes the sales scenario information. If it is a supermarket promotion scenario with a wide variety of products and rich promotional activities, the system will focus on strengthening the classification display function of the product display module. Using big data analysis technology, it mines historical sales data and divides products into different areas such as popular product areas, promotional product areas, and associated product recommendation areas for display according to factors such as product sales popularity and associated purchase situations. For example, in the promotional product area, products participating in full reduction and buy-one-get-one-free activities are centrally displayed, and the activity content is highlighted through eye-catching signs.

[0039] For the payment method information, if it is determined that multiple payment methods such as face recognition payment, QR code payment, and bank card payment are supported, the payment processing module will make corresponding configurations for each payment method. Taking face recognition payment as an example, it deeply connects with professional face recognition payment service providers to ensure the security and stability of the payment process. Using advanced 3D face recognition technology, compared with traditional 2D face recognition, it can more effectively resist fraud means such as photos and videos. At the same time, it optimizes the payment process and reduces the waiting time for face recognition payment. Through hardware acceleration and algorithm optimization, the recognition time of face recognition payment is shortened to within 1 second. For QR code payment, the system automatically adapts to the QR code formats of multiple mainstream payment platforms to ensure the accuracy and speed of QR code recognition.

[0040] In terms of commodity pricing methods, if there are multiple methods such as pricing by weight, pricing by piece, combined pricing, etc., the pricing calculation module will use different algorithms for price calculation. For commodities priced by weight, the system conducts real-time data interaction with a high-precision electronic scale to ensure accurate acquisition of weight data and calculates based on the preset unit price. For commodities with combined pricing, such as the commodity combination of "buy two get one free", the system automatically identifies the commodity combination through a rule engine and accurately calculates the actual price to be paid.

[0041] To improve the scalability and compatibility of the system, a microservices architecture is adopted to build the infrastructure of the self-checkout system. The commodity display module, payment processing module, and pricing calculation module are split into independent microservices, and each microservice can be developed, deployed, and upgraded independently without affecting each other. In this way, when the system needs to add new payment methods or pricing methods, only the corresponding microservices need to be modified and updated, rather than making large-scale changes to the entire system.

[0042] In some embodiments, during the operation of the self-checkout system, the effective management and display of commodities are crucial for enhancing the shopping experience and promoting sales. The system will perform a series of operations related to commodity management and display before determining the infrastructure of the self-checkout system. Specifically, the system uses a deep learning model to process the commodity photo information. In this process, the model can accurately extract rich feature information from the commodity photos to form a commodity feature information set, which at least includes the appearance features, packaging features, and brand logo features of the commodity. The appearance features include the shape, color, size, etc. of the commodity; the packaging features include details such as the material, pattern, and text of the commodity packaging; the brand logo features include iconic elements such as the brand name and trademark of the commodity. Then, combining the extracted commodity feature information set and the preset commodity classification rules, the system classifies the commodities and generates commodity classification labels. The preset commodity classification rules are preset based on various factors such as the attributes, uses, and sales characteristics of the commodities. Specifically, classification based on attributes includes classifying commodities into major categories such as food, daily necessities, and electronic products. Under the food category, it can be further subdivided, such as snacks, beverages, fresh food, etc. If the commodity features show that a commodity has the appearance and packaging features of food and contains food-related ingredient information, the system will classify it under the food category and generate the corresponding classification label; classification based on uses includes classifying commodities into toiletries, office supplies, kitchen supplies, etc.; classification based on sales characteristics includes classifying them into popular commodities, promotional commodities, new products, etc.

[0043] Based on the product classification labels and current sales scenario information, the system generates a product display plan. Different sales scenarios have different requirements for product display, and the system will comprehensively consider these factors to formulate the plan. According to sales data and market demand, the system determines the key recommendation areas for products. For new products or high-profit products, they will be placed in the key recommendation areas and attract customers' attention through special display methods and signs. Based on the generated product display plan, the system dynamically adjusts the display logic of the product display module in the basic infrastructure of the self-service checkout system. The product display module is responsible for displaying product information on the checkout display terminal or physical shelves. On the checkout display terminal, the system adjusts the display order and method of products according to the product display plan, placing the key recommended products at the top or center of the page to highlight them. At the same time, according to the product classification labels, the same category of products are displayed together for the convenience of customers to find. For example, when displaying food products, sub-categories such as snacks and beverages will be listed separately, and the products under each category are sorted and displayed according to sales volume or promotion intensity. The self-service checkout system can also be associated with physical shelves. The system adjusts the actual placement position of products on the physical shelves through interaction with the shelf management system, and guides staff to place products in the correct position through electronic tags or intelligent shelf devices and adjusts them in real time according to sales situations. For example, when the sales volume of a certain product suddenly increases, the system will prompt the staff to adjust the product from the ordinary shelf position to a more prominent position to meet customers' purchase needs. Through this series of operations, the self-service checkout system can achieve intelligent classification, accurate recommendation and dynamic display of products, improve the product display effect and user shopping experience, and promote product sales.

[0044] In some embodiments, in today's complex and ever-changing market environment, the dynamic adjustment of commodity prices is crucial for merchants' business decisions and profitability. To achieve more accurate and flexible commodity pricing, in some embodiments, the self-checkout system takes a series of measures based on data analysis and model prediction. Specifically, the system can build a model that can accurately predict commodity prices. The system will collect a large amount of historical transaction data, which includes detailed information such as the selling price, sales volume, and sales time of commodities over a past period. At the same time, it will also collect seasonal factor data, such as the impact of different seasons and holidays on commodity demand and prices. For example, the price and sales volume of cold drinks usually rise in summer, while they may decline in winter; during the Spring Festival, there will be obvious changes in the price and demand of gift commodities. Then, deep learning algorithms are used to train the collected data. These algorithms can process sequential data and well capture the changing patterns of prices over time and the impact of seasonal factors. During the training process, the model will continuously adjust its own parameters to minimize the error between the predicted price and the actual price. After repeated training with a large amount of data, the model gradually learns the complex relationship between historical transaction data, seasonal factors, and commodity prices, thus acquiring the ability to predict commodity prices.

[0045] After the model is built, to ensure the timeliness and accuracy of price prediction, the system needs to obtain the latest price information in the market in real time. Specifically, the system collects market price data in real time through various channels, which can include the websites of competitors, industry price monitoring platforms, price information provided by suppliers, etc. The collected market price data is input into the trained commodity price prediction model. The model will predict the future price trend of the commodity based on the input data and the previously learned patterns, and obtain the commodity price prediction result. This result can be a specific price value or a trend judgment of price increase or decrease.

[0046] After obtaining the predicted results of commodity prices, the system will automatically adjust the commodity pricing strategy in the pricing calculation module in combination with the current sales scenario information. Based on the predicted results of commodity prices and sales scenario information, the system will automatically adjust the commodity pricing strategy in the pricing calculation module. If the predicted results show that the market price will rise and the current scenario is a non-promotional one, the system may appropriately increase the selling price of the commodity; if the predicted price drops and there is a current promotional activity, the system may further increase the price cut to enhance the competitiveness of the commodity. For example, when it is predicted that the market price of a certain brand of mobile phone is about to decline and the store is conducting a membership-exclusive activity, the system will automatically adjust the pricing strategy of the mobile phone to offer more favorable prices to members. By constructing a commodity price prediction model, collecting market price data in real time, and adjusting the commodity pricing strategy according to the predicted results, the self-checkout system can achieve dynamic management of commodity prices, helping merchants better respond to market changes and improve operational efficiency.

[0047] S105. After receiving the user instruction sent by the cashier display terminal, modify the interface style, color scheme, and operation button layout of the cashier display terminal in combination with the basic architecture of this self-checkout system.

[0048] After the self-checkout system of the self-checkout system receives the user instruction sent by the cashier display terminal, it first parses the instruction. The instruction parsing module uses natural language processing technology and instruction format recognition algorithms to accurately judge the user's intention, whether it is to adjust the color of the interface, the button layout, or perform other personalized settings.

[0049] If the user hopes to modify the interface style, the system will select the corresponding style template from the pre-designed interface style library. The style library contains various style templates, such as simple modern style, cute cartoon style, business grand style, etc., to meet the aesthetic needs of different users. According to the user instruction, the system applies the selected style template to the cashier display terminal. During the application process, the dynamic style sheet technology (CSS) is used to uniformly modify elements such as the font, icons, and background of the interface. For example, when the user selects the simple modern style, the system will adjust the interface font to a simple sans-serif font, replace the icons with flat icons with simple lines, and set the background color to a light color system to create a simple and efficient visual effect.

[0050] When it comes to color scheme adjustment, the system provides a color picker tool. Users can click on the color picker to select their favorite colors from the color wheel or preset color combinations. The system will preview the color effects selected by the users in real time and apply the new color configuration to various elements of the interface. To ensure the rationality and visual comfort of the color scheme, the system also incorporates a color matching algorithm. Based on the principles of color science, it provides some recommended color combinations for users. For example, when a user selects a primary color, the system will automatically recommend complementary and accent colors to ensure the overall color harmony of the interface.

[0051] For the modification of the operation button layout, the system uses a layout algorithm for adjustment. If a user finds the position of a certain button inconvenient to operate, the system can, according to the user's instructions, move the button to a more appropriate position. The layout algorithm takes into account factors such as the usage frequency, importance of the button, and the user's operation habits to rearrange the buttons. For example, the commonly used "Settlement" button is placed in a prominent and easily clickable position at the bottom of the interface, and the less frequently used "Print Receipt" button is adjusted to a relatively less important position.

[0052] In addition, the system also supports the multi-language interface switching function. According to the region where the store is located or the user's selection, the system can switch the interface language of the cash register display terminal to the corresponding language. This function is realized through language resource files and an internationalization framework to ensure that the system can accurately display text information in different languages, enhancing the versatility and scope of application of the system.

[0053] In the embodiments of this application, since a sales scenario recognition model based on deep learning is used to analyze retail scene photos, the payment method is flexibly determined based on the hardware device information, the pricing method is accurately determined by combining the scenario and commodity information, and a customizable self-checkout system infrastructure is built. At the same time, it supports flexibly adjusting the interface of the cash register display terminal according to the user's instructions. Therefore, it can enable the self-checkout system to automatically adapt to the diverse needs of different retail scenarios, effectively solve the problems in the prior art that the self-checkout system is difficult to quickly adapt to business changes and usage scenarios, and the system functions do not match the actual scenarios. Furthermore, it realizes the scenario-adaptive development of the self-checkout system, significantly improves the system development efficiency, reduces the development cost, ensures the practicality and usability of the system in different scenarios, and enhances the user experience and checkout efficiency.

[0054] In some embodiments, to enhance the user experience of the self-checkout system and make it more in line with the usage habits and needs of users, the system deeply analyzes the operation behavior data of users with the help of machine learning algorithms, constructs an evaluation model based on this, optimizes the interface design, and realizes personalized interface adaptation. The system uses technical means to collect the operation behavior data of users during the use of the self-checkout system. These data include key information such as click hotspots, stay duration, and operation paths. Click hotspots reflect the areas where users frequently click on the checkout display terminal, revealing users' attention to different function modules; the stay duration reflects the time users stay on specific interface elements and can be used to judge the ease of understanding information or operation for users; the operation path shows the interfaces and the order of operations that users visit in sequence during the transaction process, helping to understand users' operation habits. Based on the collected operation behavior data, the system uses machine learning algorithms to construct a user experience evaluation model. This model aims to evaluate the usability indicators of interface interaction in real time, such as the clarity of the interface, the convenience of operation, and the readability of information. Through learning a large amount of operation behavior data, the model can discover the rules and patterns behind the data, and then accurately judge the user experience when using the interface. For example, if it is found that users stay for a long time at a certain operation step and the click hotspots of this step are relatively scattered, the model may judge that this operation process is not clear enough and affects the user experience. According to the usability indicators obtained from the user experience evaluation model, the system optimizes the interface layout and interaction process. If the evaluation result shows that the position of a certain function button is not convenient for users to click, the system will adjust it to a position that is more in line with users' operation habits; if it is found that the information on a certain interface is too complex, resulting in users staying for too long, the system will simplify the design of this interface and highlight the key information.

[0055] Through these optimization measures, a personalized interface adaptation solution is generated to meet the needs of different users. The system uses camera recognition technology to detect the user group currently using the self-checkout system. When a set user group is detected, the checkout display terminal is automatically switched to the corresponding interface adaptation solution. For example, for the elderly user group, the system will switch to an adaptation solution with larger interface fonts, higher color contrast, and a more simplified operation process to facilitate the operation of elderly users; for the young user group who often use the self-checkout system, a personalized interface with richer functions and more efficient operations can be provided. Through the above steps, the self-checkout system can realize the intelligent analysis of users' operation behaviors, continuously optimize the interface design, provide personalized interaction experiences for different user groups, and improve user satisfaction and the usage efficiency of the self-checkout system.

[0056] After combining the above content, the following is a further and more specific process description of the method provided in this embodiment. Please refer to Figure 2, which is another process schematic diagram of the adaptive rapid development self-checkout method in the embodiments of this application.

[0057] S201. Determine business rule information according to sales scenario information and commodity pricing methods. The business rule information includes at least promotion activity rules, membership point rules, and inventory management rules. After the self-checkout system obtains the sales scenario information and commodity pricing methods, it will use big data analysis and intelligent algorithms to determine the business rule information. The system first classifies the sales scenarios, such as dividing them into daily sales, holiday promotions, member-exclusive activities, etc. For different sales scenarios, corresponding business rules are formulated in combination with the commodity pricing methods.

[0058] In terms of formulating promotion activity rules, the system will match corresponding promotion templates from the promotion activity database. The template contains information such as the start and end times of the promotion activity, the scope of participating commodities, and the form of discounts. At the same time, the system uses data analysis tools to mine historical sales data, analyze the sales trends and customer purchase behaviors of such commodities in past promotion activities, and optimize the promotion activity rules based on this. For example, it is found that in the past holiday promotions of a certain brand of snacks, the "buy three get one free" activity has a significant effect on increasing sales, but if combined with a full reduction activity, the sales volume will further increase. Therefore, the promotion activity rules generated by the system may be "when purchasing designated brand snacks, subtract 10 yuan for every 50 yuan spent, and buy three get one free".

[0059] For membership point rules, the system will comprehensively consider the sales scenario and commodity profit to determine the point acquisition ratio. In the daily sales scenario, for ordinary commodities priced by unit price, 1 point can be obtained for every 1 yuan of consumption; but in the member-exclusive activity scenario, for high-profit commodities, such as imported beauty products, 3 points can be obtained for every 1 yuan of consumption. The system will also set the validity period and redemption rules of the points, and determine a reasonable points redemption commodity or coupon plan through the analysis of member consumption behaviors and point usage data.

[0060] In determining the inventory management rules, the system combines the sales scenario and the sales speed of the commodity to set the inventory warning value and replenishment strategy. For commodities with large sales volumes in the promotion scenario, such as beverages during the summer promotion period, the system will increase the inventory warning value, and trigger an alarm when the inventory is lower than 100 pieces, reminding the merchant to replenish the stock in time.

[0061] To make the business rules more intelligent and accurate, the system introduces machine learning algorithms. By continuously collecting and analyzing sales data, customer behavior data, and market dynamics data, the machine learning model can automatically adjust the business rules. For example, when the system finds that the demand for a certain type of commodity in a specific region suddenly increases in a specific season, the model will automatically increase the inventory warning value of the commodity in that region and adjust the promotion activity rules to increase the promotion intensity of the commodity.

[0062] S202. Use natural language processing technology to collect and analyze the information set sent from the cash register display terminal. The information set includes the evaluation information, consultation information input by the user, and voice interaction information in the retail scenario. The self-service cash register system collects the evaluation information, consultation information input by the user, and voice interaction information in the retail scenario by setting input boxes, voice collection devices, etc. on the cash register display terminal. After the user completes the shopping settlement, the system will prompt the user to make evaluations and feedback. The user can enter a text evaluation in the input box or give feedback by voice.

[0063] The system uses the speech recognition technology in natural language processing (NLP) technology to convert the user's voice interaction information into text information. In order to deeply understand the user's intention and emotion, the system adopts sentiment analysis technology. The sentiment analysis model is based on the neural network architecture of deep learning and is trained on a large amount of text data with sentiment annotations, and can identify positive, negative or neutral sentiment in the text. For example, when the user evaluates "This product is great and I like it very much", the sentiment analysis model can judge that this evaluation is a positive sentiment; while for "This product is too difficult to use and I will never buy it again", the model can identify a negative sentiment.

[0064] When analyzing the consultation information, the system uses information retrieval technology to search for relevant answer information from the pre-constructed knowledge base. The knowledge base stores the answers to common questions, product information, promotion activity rules, etc. When the user consults "What are the preferential activities for this product?", the system extracts keywords and understands semantics, finds the corresponding preferential activity information from the knowledge base and feeds it back to the user.

[0065] S203. Combine the current sales scenario information and the commodity pricing method to perform sentiment analysis and intention recognition on the information set. In terms of sentiment analysis, the system will make a more detailed interpretation of the sentiment analysis results according to different sales scenarios. In the promotion scenario, the user's evaluation and feedback may be affected by the promotion activities. If the user evaluates "The price is very cost-effective" after purchasing a product participating in the full reduction activity, the system will not only identify this as a positive sentiment, but also analyze the user's recognition of the price discount in combination with the promotion scenario. At this time, the system can feed this information back to the merchant and suggest that the merchant continue to adopt a similar full reduction strategy in subsequent promotion activities.

[0066] For the commodity pricing method, the system will also consider it in sentiment analysis. If it is a fresh food commodity priced by weight and the user comments that "the weighing process is very troublesome and I hope there can be a more convenient way", the system can identify the user's dissatisfaction with the pricing method of this commodity. In response to this situation, the system can provide suggestions for merchants, such as optimizing the weighing process or considering pricing the commodity by portion during specific time periods to improve the user experience.

[0067] In terms of intent recognition, the system utilizes semantic understanding and knowledge graph technology. The knowledge graph contains knowledge in multiple aspects such as commodity information, sales scenarios, and user behavior. By conducting correlation analysis between the user's feedback information and the knowledge graph, the system can more accurately identify the user's intent. For example, when the user asks "Are there any commodities that match this one?", combined with the current sales scenario and commodity information, the system can identify that the user has the intent to purchase related commodities. At this time, the system can screen out related commodities from the commodity database and recommend them to the user at the cash register display terminal. For example, when purchasing bread, milk is recommended. The system continuously interacts with the user and adjusts the intent recognition strategy according to the user's feedback. When the system successfully identifies the user's intent and meets the user's needs, it will receive a positive feedback reward; if the recognition is incorrect, it will receive a negative feedback penalty. In this way, the system's intent recognition ability will continuously improve and be able to more accurately capture the user's potential needs.

[0068] S204: If it is recognized that the customer has a set demand for the target commodity, then feedback it to the merchant management terminal according to the set demand; When the self-service cash register system determines through natural language processing technology, sentiment analysis, and intent recognition that the customer has a set demand for the target commodity, it enters the link of feedbacking information to the merchant management terminal. The system first structurally processes the recognized customer demand and organizes the demand information according to a preset format. Then, it uses a secure and reliable encryption communication protocol to dock with the merchant management terminal through an interface. The merchant management system only needs to be developed according to the interface specifications to easily access the self-service cash register system and receive customer demand information.

[0069] S205: After receiving the adjustment information sent by the merchant management terminal, adjust the business rule information in real time according to the adjustment information.

[0070] Merchants can view the detailed information of customer demands, including statistical data such as the specific content of the demands, the number of customers who put forward the demands, and the frequency of the demands. Through this interface, merchants can quickly locate the products and demands that customers pay more attention to, and after determining the adjustment information, the merchants can send it to the self-checkout system through the merchant management terminal. The system uses an information parsing module to parse the adjustment information sent from the merchant management terminal. Based on the preset information format specification, this module can accurately identify and adjust each parameter in the adjustment information, such as the type of business rule to be adjusted (promotion activity rule, membership point rule, or inventory management rule), the specific adjustment content (such as the change in the preferential intensity of the promotion activity, the adjustment of the membership point acquisition ratio, the change in the inventory warning value), etc. At the same time, the system also has a version management function. Each adjustment of the business rule will be recorded to form a version history. Merchants can view the change records of the business rules at any time, compare the impacts of different versions of the rules on indicators such as sales performance and customer satisfaction, and provide a reference basis for subsequent rule adjustments.

[0071] In the embodiments of the present application, due to the adoption of technologies such as big data analysis, intelligent algorithms, and natural language processing, the business rule information is determined based on the sales scenario and the commodity pricing method, the user feedback is deeply analyzed using natural language processing technology, the business rules are adjusted according to the user demands, and the rules are updated in real time in combination with the merchant adjustment information. At the same time, it has a version management function, so it can realize the intelligent and dynamic management of business rules, effectively solve the problems in the prior art that the business rules of the self-checkout system are difficult to flexibly adjust, cannot accurately respond to user demands and market changes, and thus improve the system operation efficiency.

[0072] The self-checkout system in the embodiments of the present invention will be described from the perspective of hardware processing below. Please refer to Figure 3 , which is a schematic structural diagram of an entity device of the self-checkout system in the embodiments of the present application.

[0073] It should be noted that Figure 3 the structure of the self-checkout system shown is only an example and should not bring any limitations to the functions and usage scopes of the embodiments of the present invention.

[0074] Such as Figure 3As shown, the self-service checkout system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes according to a program stored in a Read-Only Memory (ROM) 302 or a program loaded from a storage section 308 into a Random Access Memory (RAM) 303, such as executing the method described in the above embodiments. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0075] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, button switches, etc.; an output section 307 including a Liquid Crystal Display (LCD), an audio output device, indicator lights, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read from it can be installed into the storage section 308 as needed.

[0076] Specifically, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network via the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by the Central Processing Unit (CPU) 301, various functions defined in the present invention are executed.

[0077] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fibers, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0078] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings.

[0079] Specifically, the self-checkout system of this embodiment includes a processor and a memory. A computer program is stored on the memory. When the computer program is executed by the processor, it implements the adaptive rapid development self-checkout method provided in the above embodiment.

[0080] On the other hand, the present invention also provides a computer-readable storage medium. This storage medium may be included in the self-checkout system described in the above embodiment; or it may exist separately and not be assembled into the self-checkout system. The above storage medium carries one or more computer programs. When the above one or more computer programs are executed by a processor of the self-checkout system, the self-checkout system is enabled to implement the adaptive rapid development self-checkout method provided in the above embodiment.

[0081] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present application.

[0082] As used in the foregoing embodiments, depending on the context, the term "when" may be interpreted to mean "if" or "after" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "when determining" or "if (the stated condition or event) is detected" may be interpreted to mean "if determined" or "in response to determining" or "when (the stated condition or event) is detected" or "in response to detecting (the stated condition or event)".

[0083] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the foregoing embodiments can be implemented. The processes can be completed by relevant hardware instructed by a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the foregoing method embodiments. The foregoing storage media include: various media such as ROM or random access memory RAM, magnetic disks, or optical discs that can store program codes.

Claims

1. An adaptive rapid development self-checkout method, applied to a self-checkout system, characterized in that The method includes: obtaining information of multiple product photos and information of retail scene photos; Based on the information of multiple product photos and the information of the retail scene photos, determining the current sales scene information in combination with a sales scene recognition model, where the sales scene recognition model is pre-trained through deep learning based on multiple labeled sales scene photos; After obtaining information of multiple hardware devices, determining the payment method information of the product in combination with the information of multiple hardware devices, where the payment method information at least includes the face recognition payment method; Determining the product pricing method in combination with the current sales scene information and the product photo information; Determining the basic architecture of the self-service cash register system based on the current sales scene information, the payment method information, and the product pricing method, where the basic architecture of the self-service cash register system at least includes a product display module, a payment processing module, and a pricing calculation module; After receiving a user instruction sent from the cash register display terminal, modifying the interface style, color combination, and operation button layout of the cash register display terminal in combination with the basic architecture of the self-service cash register system.

2. The method according to claim 1, wherein After the step of determining the basic architecture of the self-service cash register system based on the current sales scene information, the payment method information, and the product pricing method, it further includes: Determining business rule information based on the sales scene information and the product pricing method, where the business rule information at least includes promotion activity rules, membership point rules, and inventory management rules; After receiving an instruction from the user to modify the rules, updating the business rule information.

3. The method according to claim 2, wherein After the step of determining business rule information based on the sales scene information and the product pricing method, it further includes: Collecting and analyzing an information set sent from the cash register display terminal by using natural language processing technology, where the information set includes evaluation information, consultation information input by the user, and voice interaction information in the retail scene; Performing sentiment analysis and intention recognition on the information set in combination with the current sales scene information and the product pricing method; If it is recognized that the customer has a set demand for the target product, then feeding back to the merchant management terminal according to the set demand; After receiving the adjustment information sent from the merchant management terminal, adjusting the business rule information in real time according to the adjustment information.

4. The method according to claim 1, wherein Before the step of determining the basic architecture of the self-service cash register system based on the current sales scene information, the payment method information, and the product pricing method, it further includes: Extracting product features from the product photo information based on a deep learning model to obtain a set of product feature information, where the set of product feature information at least includes product appearance features, packaging features, and brand logo features; Classifying the products in combination with the set of product feature information and a preset product classification rule to generate product classification labels; Generating a product display plan based on the product classification labels and the current sales scene information, where the product display plan includes product placement positions, product combination displays, and product key recommendation areas; Dynamically adjusting the display logic of the product display module in the basic architecture of the self-service cash register system based on the product display plan.

5. The method according to claim 1, wherein Before the step of determining the basic architecture of the self-service checkout system according to the current sales scenario information, the payment method information, and the commodity pricing method, the following steps are further included: Construct a commodity price prediction model, which is trained based on a deep learning algorithm by analyzing historical transaction data and seasonal factor data; Collect market price data in real time, input the market price data into the commodity price prediction model, and obtain a commodity price prediction result; According to the commodity price prediction result, automatically adjust the commodity pricing strategy in the pricing calculation module in combination with the current sales scenario information.

6. The method according to claim 1, wherein After the step of modifying the interface style, color matching, and operation button layout of the cashier display terminal in combination with the basic architecture of the self-service checkout system after receiving a user instruction sent by the cashier display terminal, the following steps are further included: Use a machine learning algorithm to analyze the operation behavior data of the user, where the operation behavior data includes click hotspots, dwell time, and operation paths; Construct a user experience evaluation model based on the operation behavior data, and the user experience evaluation model is used to evaluate the usability indicators of interface interaction in real time; Optimize the interface layout and interaction process according to the usability indicators, and generate a personalized interface adaptation plan; When a set user group is detected through a camera, control the cashier display terminal to automatically switch to the corresponding interface adaptation plan.

7. The method according to claim 1, characterized in that, The step of determining the current sales scenario information according to the multiple commodity photo information and the retail scenario photo information in combination with the sales scenario recognition model specifically includes: While obtaining the multiple commodity photo information and the retail scenario photo information through a camera, obtain environmental illuminance data and passenger flow data through a sensor; Establish a scene feature vector in combination with the environmental illuminance data and the passenger flow data; Input the scene feature vector into the sales scenario recognition model, and output a scene classification result; Determine the current sales scenario information according to the scene classification result.

8. A self-checkout system, characterized in that, The self-service checkout system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the self-service checkout system to execute the method according to any one of claims 1-7.

9. A computer-readable storage medium, comprising instructions, characterized in that, When the instruction runs on the self-service checkout system, enable the self-service checkout system to execute the method according to any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product runs on the self-service checkout system, enable the self-service checkout system to execute the method according to any one of claims 1-7.

Images