Weighing settlement method and system based on intelligent electronic scale

Abstract

The invention discloses a weighing settlement method and system based on an intelligent electronic scale, and the method comprises the steps: collecting a user image, and the initial image information and initial weight information of a to-be-weighed and settled article when it is detected that the to-be-weighed and settled article is placed on a scale pan of the intelligent electronic scale; processing the initial image information and the initial weight information to obtain target image information and target weight information; performing image recognition on the target image information to obtain a recognition result; if the identification result is that at least two similar candidate items exist, obtaining user information and historical purchase data associated with the user image; determining a target article from the at least two similar candidate articles according to the user information, the historical purchase data, the identification result and the target weight information; and obtaining unit price information of the target article, and determining settlement information of the to-be-weighed and settled article according to the unit price information and the target weight information. According to the invention, misrecognition of similar articles is effectively reduced, and the accuracy of weighing settlement is improved.

Description

Technical Field

[0001] This invention relates to the field of electronic scale technology, and in particular to a weighing and settlement method and system based on an intelligent electronic scale. Background Technology

[0002] In existing commercial retail scenarios, small and micro businesses that require weighing, such as fresh produce stores, small convenience stores, delicatessens, and farmers' market stalls, generally suffer from fragmented operations and low efficiency in the weighing and payment processes. Traditional electronic scales only have a single weighing function. After weighing, cashiers must switch to separate devices such as payment speakers and POS machines to process payments. This separation of two consecutive actions makes the operation process cumbersome and increases transaction time. Although some integrated weighing and payment machines have appeared on the market, these devices mostly integrate complex SaaS POS systems such as Android or Windows on top of electronic scales. They not only have high operating thresholds and poor portability, but also rely on WiFi networks to realize payment functions. At the same time, their procurement costs are high, making it difficult to adapt to the low-cost and easy-to-operate business needs of small and micro merchants. In addition, these devices have insufficient recognition accuracy for highly similar items (such as Northeast rice and Wuchang rice, or similar rice from different origins or brands). Differences in unit price due to differences in the origin, grade, and supplier of items can easily lead to recognition errors, further affecting the accuracy of weighing and settlement. Summary of the Invention

[0003] This invention provides a weighing and settlement method and system based on an intelligent electronic scale, aiming to solve the problem of low accuracy in weighing and settlement caused by the easy misidentification of similar items by existing electronic scales.

[0004] In a first aspect, embodiments of the present invention provide a weighing and settlement method based on an intelligent electronic scale, comprising: When an item to be weighed and settled is detected to be placed on the weighing pan of the smart electronic scale, the user's image and the initial image information and initial weight information of the item to be weighed and settled are collected. The initial image information and the initial weight information are processed to obtain the target image information and the target weight information; Image recognition is performed on the target image information to obtain the recognition result; If the identification result indicates that there are at least two similar candidate items, then obtain the user information and historical purchase data associated with the user image; The target item is determined from at least two similar candidate items based on the user information, the historical purchase data, the identification result, and the target weight information. Obtain the unit price information of the target item, and determine the settlement information of the item to be weighed and settled based on the unit price information and the target weight information.

[0005] Secondly, embodiments of the present invention also provide a weighing and settlement system based on an intelligent electronic scale, including an intelligent electronic scale. The intelligent electronic scale includes a control processing module and a weighing module, an image acquisition module, and a product recognition module connected to the control processing module. The weighing module is used to acquire initial weight information of the item to be weighed and settled through a high-precision sensor, and transmit the initial weight information to the control processing module for processing to obtain target weight information. The image acquisition module is used to acquire initial image data of a user image and the item to be weighed and settled from multiple angles, and transmit the initial image data to the control processing module for processing to obtain target image information. The product recognition module identifies the target image information to obtain a recognition result. When the recognition result indicates the existence of at least two similar candidate items, it acquires user information and historical purchase data associated with the user image, and determines the target item from the at least two similar candidate items based on the user information, the historical purchase data, the recognition result, and the target weight information. The control processing module is used to determine the settlement information of the item to be weighed and settled based on the target item and the unit price information corresponding to the target item.

[0006] This invention provides a weighing and settlement method and system based on a smart electronic scale. The method includes: when an item to be weighed and settled is detected placed on the weighing pan of the smart electronic scale, acquiring a user image and initial image and initial weight information of the item to be weighed and settled; processing the initial image and initial weight information to obtain target image and target weight information; performing image recognition on the target image information to obtain a recognition result; if the recognition result indicates the existence of at least two similar candidate items, acquiring user information and historical purchase data associated with the user image; determining the target item from the at least two similar candidate items based on the user information, the historical purchase data, the recognition result, and the target weight information; acquiring the unit price information of the target item, and determining the settlement information of the item to be weighed and settled based on the unit price information and the target weight information. The technical solution of this invention, after identifying at least two similar candidate items in an image, determines the target item from the at least two similar candidate items based on user information, historical purchase data, identification results, and target weight information. It then determines the settlement information of the item to be weighed and settled based on the target item and its unit price information, effectively reducing the misidentification of similar items and improving the accuracy of weighing and settlement. Attached Figure Description

[0007] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 A block diagram of a weighing and settlement system based on an intelligent electronic scale provided in an embodiment of the present invention; Figure 2 A block diagram of an intelligent electronic scale provided in an embodiment of the present invention; Figure 3 This is a flowchart illustrating a weighing and settlement method based on an intelligent electronic scale, as provided in an embodiment of the present invention. Figure 4 This is a schematic diagram of a sub-process of a weighing and settlement method based on an intelligent electronic scale, provided in an embodiment of the present invention. Figure 5 This is a schematic block diagram of a weighing and settlement device based on an intelligent electronic scale, provided as an embodiment of the present invention.

[0009] Figure label: 100. Weighing and settlement system based on intelligent electronic scale; 10. Intelligent electronic scale; 11. Control and processing module; 12. Weighing module; 13. Image acquisition module; 14. Commodity recognition module; 15. QR code payment module; 16. Display module; 17. Input module; 18. Voice broadcast module; 19. Communication module; 20. Cloud server; 30. Printer; 40. Traceability platform. Detailed Implementation

[0010] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0011] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0012] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0013] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0014] As used in this specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [described condition or event] is detected" may be interpreted, depending on the context, as "once determined," "in response to determination," "once [described condition or event] is detected," or "in response to detection of [described condition or event]."

[0015] This invention proposes a weighing and settlement method and system based on an intelligent electronic scale, which solves the problem of low accuracy in weighing and settlement caused by the easy misidentification of similar items in existing electronic scales. In this embodiment, after identifying at least two similar candidate items from the image, the target item is determined from the at least two similar candidate items based on user information, historical purchase data, identification results, and target weight information. The settlement information of the item to be weighed and settled is determined based on the target item and its unit price information, which effectively reduces the misidentification of similar items and improves the accuracy of weighing and settlement.

[0016] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0017] This invention proposes a weighing and settlement method based on a smart electronic scale. This method can be used in a weighing and settlement system based on a smart electronic scale, such as... Figure 1 and Figure 2As shown, the weighing and settlement system 100 based on an intelligent electronic scale includes an intelligent electronic scale 10. The intelligent electronic scale 10 includes a control processing module 11 and a weighing module 12, an image acquisition module 13, and a product recognition module 14 connected to the control processing module 11. The weighing module 12 is used to acquire the initial weight information of the item to be weighed and settled through a high-precision sensor, and transmit the initial weight information to the control processing module 11 for processing to obtain the target weight information. The image acquisition module 13 is used to acquire user images and initial image data of the item to be weighed and settled from multiple angles, and transmit the initial image data... The target image information is processed by the control processing module 11; the product recognition module 14 recognizes the target image information to obtain a recognition result. When the recognition result indicates that there are at least two similar candidate items, user information and historical purchase data associated with the user image are obtained. Based on the user information, the historical purchase data, the recognition result, and the target weight information, the target item is determined from the at least two similar candidate items. The control processing module 11 is used to determine the settlement information of the item to be weighed and settled based on the target item and the unit price information corresponding to the target item. It should be noted that, in this embodiment, as... Figure 2 As shown, the intelligent electronic scale 10 also includes a QR code payment module 15, a display module 16, an input module 17, and a voice broadcast module 18. The QR code payment module 15 is used to perform payment operations. Specifically, the QR code payment module 15 has a built-in or external high-performance QR code scanner that supports multiple mainstream QR code payment methods such as Alipay, WeChat Pay, and UnionPay QuickPass. The scanner has the ability to quickly recognize QR codes and can adapt to QR code scanning at different angles and with varying clarity. The display module 16 is used to display settlement information. The input module 17 is used to assist in input... The input module 17 is used to input or correct the product information of the items to be weighed and settled. It is understood that the input module 17 includes numeric keys, function keys, etc., for inputting the unit price of the product and setting functions (such as total amount accurate to yuan, jiao, fen, rounding down, adjusting the voice broadcast volume, etc.). The keys are waterproof and wear-resistant to ensure long-term stable use. The voice broadcast module 18 is used to broadcast the settlement information. It is understood that the voice broadcast module 18 has a built-in high-quality speaker, which can clearly and accurately broadcast the settlement information, and can also provide timely voice reminders in case of payment abnormalities.

[0018] It should also be noted that, in this embodiment, as Figure 1 As shown, the weighing and settlement system 100 based on the intelligent electronic scale also includes a cloud server 20, a printer 30, and a traceability platform 40 that are communicatively connected to the intelligent electronic scale 10; as Figure 2As shown, the intelligent electronic scale 10 also includes a communication module 19. The communication module 19 is used to send the settlement information to the printer 30 for printing, and to upload the transaction information and traceability information generated by the control processing module 11 to the cloud server 20 and the traceability platform 40, respectively. It should be further noted that the specific functions implemented in the weighing module 12, the image acquisition module 13, the product recognition module 14, the control processing module 11, the QR code payment module 15, the display module 16, the input module 17, and the voice broadcast module 18 will be described in detail in later embodiments, and will not be repeated here for simplicity.

[0019] Please refer to Figure 3 , Figure 3 A flowchart illustrating the weighing and settlement method based on an intelligent electronic scale according to an embodiment of the present invention is shown, as follows: Figure 3 As shown, the weighing and settlement method based on the intelligent electronic scale includes steps S110-S160.

[0020] S110. When an item to be weighed and settled is detected to be placed on the weighing pan of the smart electronic scale, the user image and the initial image information and initial weight information of the item to be weighed and settled are collected.

[0021] In this embodiment, when the pressure sensor of the smart electronic scale detects a change in the weight on the weighing pan and the weight value stably exceeds a preset threshold (e.g., 5g), it is determined that an item to be weighed and settled has been placed there. Then, multi-module collaborative acquisition is triggered: the front-facing facial capture camera is activated to capture the user's facial image within 0.5 seconds, obtaining the user image; simultaneously, the binocular item image capture camera is activated to perform 360° surround shooting of the weighing pan area, acquiring initial image information of the item to be weighed and settled from 3 different angles with a resolution of 1080P; at the same time, the weighing module collects the initial weight information at a frequency of 10 times / second.

[0022] S120. The initial image information and the initial weight information are processed to obtain target image information and target weight information.

[0023] In this embodiment, the initial weight information is denoised to obtain denoised weight data; When the fluctuation error of the denoised weight data for a consecutive preset number of times is less than the preset weight data fluctuation error, the denoised weight data is used as the target weight information; the target image information is obtained by normalizing the initial image data. It should be noted that in this embodiment, a Kalman filter algorithm is used to denoise the acquired original weight information. It should also be noted that in this embodiment, the normalization operation on the initial image data includes uniformly adjusting the image size of the initial image data to 224×224 pixels and normalizing the pixel values ​​to the 0-1 range to eliminate interference from image size differences.

[0024] S130, Perform image recognition on the target image information to obtain the recognition result.

[0025] In this embodiment, the target image information is input into a lightweight convolutional neural network model to extract item features and obtain item feature vectors. The item feature vectors are then compared with a preset product feature library using cosine similarity to filter out candidate items with a similarity greater than a preset similarity, thus obtaining the recognition result. It should be noted that in this embodiment, the lightweight convolutional neural network model is a model trained based on the MobileNetV3 architecture combined with data collected from the item recognition scenario. It should also be noted that in this embodiment, the item feature vectors are compared with the preset product feature library using cosine similarity, and the cosine value of the angle between the item feature vectors and the feature vectors in the preset product feature library is calculated. Based on the cosine value of the angle, candidate items with a similarity greater than a preset similarity are filtered out to obtain the recognition result.

[0026] S140. If the identification result indicates that there are at least two similar candidate items, then obtain the user information and historical purchase data associated with the user image.

[0027] In this embodiment, if the identification result indicates the existence of at least two similar candidate items, the collected user image is compared with a preset user profile database to obtain associated user information; simultaneously, the user's historical purchase data for the past three months is retrieved from the transaction record database. It should be noted that in this embodiment, if the identification result shows only one candidate item, that candidate item is directly used as the target item.

[0028] S150. Determine the target item from at least two similar candidate items based on the user information, the historical purchase data, the identification result, and the target weight information.

[0029] In this embodiment, as Figure 4As shown, step S150 specifically includes steps S151-S154: S151, determining the weight weight coefficient of the candidate item based on historical purchase data, the identification result, and the target weight information; S152, determining the user purchase weight coefficient of the candidate item based on the user information, and determining the purchase frequency coefficient of the candidate item based on the historical purchase data; S153, calculating the comprehensive score of the candidate item based on the weight weight coefficient, the user purchase weight coefficient, and the purchase frequency coefficient; S154, selecting the candidate item with the highest comprehensive score as the target item. It should be noted that in this embodiment, step S151 includes: obtaining the historical purchase weight within a preset time period corresponding to the candidate item from the historical purchase data based on the identification result; calculating the historical average weight of the candidate item based on the historical purchase weight of each candidate item; calculating the proximity of the candidate item based on the target weight information and the historical average weight of each candidate item; and assigning the weight weight coefficient to the candidate item with the highest proximity. Specifically, for ease of understanding, let's assume the identification results are similar candidate items "Northeast Rice" and "Wuchang Rice", and the target weight information is 2.5kg. Based on the identification results, the historical purchase weights of the two candidate items within the past three months are obtained from the historical purchase data: Northeast Rice is [2.4, 2.6, 2.5, 2.5] kg, and Wuchang Rice is [1.0, 1.2, 0.9, 1.1] kg; the historical average weights are calculated: Northeast Rice (2.4+2.6+2.5+2.5) / 4=2.5 kg, Wuchang Rice (1.0+1.2+0.9+1.1) / 4=1.05 kg; the proximity is calculated using the formula "proximity = 1 - |target weight information - historical average weight| / historical average weight": Northeast Rice proximity = 1 - |2.5-2.5| / 2.5=1, Wuchang Rice proximity = 1 - |2.5-1.05| / 1.05≈-0.38; Northeast Rice, with the highest proximity, is assigned a weight weight coefficient of 0.2.

[0030] In one embodiment, such as this embodiment, step 152 specifically includes: classifying the age information in the user information into youth, middle-aged, and elderly; calculating the historical average order value based on the consumption amount and purchase frequency in the user information, and classifying the user's consumption habits into high-end, mid-range, and low-end based on the historical average order value; determining the user purchase weight coefficient of the candidate item based on the age information and the user's consumption habits; and determining the purchase frequency coefficient based on the purchase frequency of the candidate item in the historical purchase data. It should be noted that in this embodiment, if the user's age is 18-35 years old, they are classified as youth; if the user's age is 36-55 years old, they are classified as middle-aged; and if the user's age is 56 years old or older, they are classified as elderly. It should also be noted that in this embodiment, if the historical average order value is ≥80 yuan, the user's consumption habits are classified as high-end; if the historical average order value is 30-79 yuan, the user's consumption habits are classified as mid-range; and if the historical average order value is <30 yuan, the user's consumption habits are classified as low-end. For example, a user with a total spending of 600 yuan and 12 purchases in the past 3 months, with an average order value of 50 yuan, is classified as mid-range; a user with a total spending of 960 yuan and 12 purchases, with an average order value of 80 yuan, is classified as high-end. This method quantifies and categorizes user spending levels. Specifically, for ease of understanding, let's assume the user's age is 32 years old, classifying them as young; their historical spending is 600 yuan, with 12 purchases, and the calculated average order value ("historical average order value = total spending / number of purchases") is 50 yuan, classifying them as having mid-range spending habits. The candidate items are "Northeast Rice" and "Wuchang Rice." Young mid-range users have a high preference for Northeast Rice, so Northeast Rice is assigned a purchase weight coefficient of 0.12. In the historical purchase data, Northeast Rice was purchased 8 times, and Wuchang Rice was purchased 2 times. Calculated using "purchase frequency coefficient = number of purchases per item / total number of purchases": the purchase frequency coefficient for Northeast Rice = 8 / 10 = 0.8, and the purchase frequency coefficient for Wuchang Rice = 2 / 10 = 0.2.

[0031] S160. Obtain the unit price information of the target item, and determine the settlement information of the item to be weighed and settled based on the unit price information and the target weight information.

[0032] In this embodiment, after determining the target item, the unit price information of the target item is obtained from the product information database; the settlement amount is calculated based on the unit price information and the target weight information, with the formula "settlement amount = unit price information × target weight information", and settlement information containing product name, unit price, weight and amount is generated.

[0033] In one embodiment, such as this embodiment, after step S160, the method further includes: displaying the settlement information and verbally announcing the total price information in the settlement information, while simultaneously displaying the payment QR code; when payment information is received, verbally announcing the payment status and payment amount; transmitting the settlement information and payment information to a printer for printing; generating transaction information and traceability information based on the settlement information, payment information, and payment method during the transaction process; uploading the transaction data to the management platform of the cloud server and uploading the traceability information to the product traceability platform. It should be noted that in this embodiment, voice announcement enhances the convenience of interaction; real-time feedback on payment status ensures efficient transaction confirmation; simultaneous printing and cloud uploading of transaction data enables transaction traceability and convenient management; and the traceability information is integrated with the product traceability platform to improve product traceability capabilities, thereby optimizing the overall transaction process and information management level for micro and small merchants.

[0034] In one embodiment, such as this embodiment, before step S160, the method further includes: displaying the target product; if a preset modification instruction is received within a preset time, then the candidate item selected by the user is obtained according to the preset modification instruction, and the user purchase weight coefficient of the candidate item selected by the user is adjusted. It should be noted that in this embodiment, the target product name, unit price, weight, and other information are displayed in sections on the smart electronic scale display screen; if a preset modification instruction is received from the user via touch button or voice within a preset time (e.g., 5 seconds), the candidate item list is immediately retrieved for the user to choose from; according to the candidate item selected by the user, the purchase weight coefficient corresponding to the candidate item is increased according to a preset rule (e.g., increased by 0.05-0.1).

[0035] Figure 5 This is a schematic block diagram of a weighing and settlement device 200 based on an intelligent electronic scale provided in an embodiment of the present invention. Figure 5 As shown, this corresponds to the above-described weighing and settlement method based on a smart electronic scale. The smart electronic scale-based weighing and settlement device 200 includes a unit for executing the above-described weighing and settlement method based on a smart electronic scale. Specifically, please refer to... Figure 5 The weighing and settlement device 200 based on the intelligent electronic scale includes a detection and acquisition unit 201, a processing unit 202, an identification unit 203, an acquisition unit 204, a first determination unit 205, and a second determination unit 206. Detailed descriptions of each functional module are as follows: The detection and acquisition unit 201 is used to acquire the user image and the initial image information and initial weight information of the item to be weighed and settled when it is detected that an item to be weighed and settled is placed on the weighing pan of the smart electronic scale. Processing unit 202 is used to process the initial image information and the initial weight information to obtain target image information and target weight information; The recognition unit 203 is used to perform image recognition on the target image information to obtain a recognition result; The acquisition unit 204 is used to acquire user information and historical purchase data associated with the user image if the recognition result indicates that there are at least two similar candidate items. The first determining unit 205 is used to determine the target item from at least two similar candidate items based on the user information, the historical purchase data, the identification result and the target weight information; The second determining unit 206 is used to obtain the unit price information of the target item and determine the settlement information of the item to be weighed and settled based on the unit price information and the target weight information.

[0036] In some embodiments, such as this one, the processing unit 202 is specifically used for: The initial weight information is denoised to obtain denoised weight data; When the fluctuation error of the denoised weight data for a consecutive preset number of times is less than the preset weight data fluctuation error, the denoised weight data is used as the target weight information. The target image information is obtained by normalizing the initial image data.

[0037] In some embodiments, such as this one, the identification unit 203 is specifically used for: The target image information is input into a lightweight convolutional neural network model to extract item features and obtain item feature vectors. The item feature vector is compared with a preset product feature library using cosine similarity to filter out candidate items with a similarity greater than a preset similarity to obtain the recognition result.

[0038] In some embodiments, such as this embodiment, the first determining unit 205 is specifically used for: The weight weight coefficient of the candidate item is determined based on historical purchase data, the identification results, and the target weight information; The user purchase weight coefficient of the candidate item is determined based on the user information, and the purchase frequency coefficient of the candidate item is determined based on the historical purchase data. The overall score of the candidate item is calculated based on the weight weight coefficient, the user purchase weight coefficient, and the purchase frequency coefficient. The candidate item with the highest overall score is selected as the target item.

[0039] In some embodiments, such as this one, the identification unit 203 is further configured to: Based on the identification result, the historical purchase weight within a preset time period corresponding to the candidate item is obtained from the historical purchase data; Calculate the historical average weight of each candidate item based on its historical purchase weight; The proximity of the candidate items is calculated based on the target weight information and the historical average weight of each candidate item. Assign the weight weight coefficient to the candidate item with the highest proximity.

[0040] In some embodiments, such as this one, the identification unit 203 is further configured to: The age information in the user information is divided into youth, middle-aged, and elderly; The historical average order value is calculated based on the consumption amount and purchase frequency in the user information, and the user's consumption habits are divided into high-end, mid-range and low-end based on the historical average order value; The user purchase weight coefficient of the candidate items is determined based on the age information and the user's consumption habits; The purchase frequency coefficient is determined based on the number of times the candidate items were purchased in the historical purchase data.

[0041] In some embodiments, such as this one, the weighing and settlement device 200 based on the smart electronic scale further includes: The display unit is used to display the settlement information, announce the total price information in the settlement information via voice, and display the payment QR code. The receiving and playback unit is used to verbally play the payment status and amount when payment information is received. A transmission unit is used to transmit the settlement information and the payment information to a printer for printing. The generation unit is used to generate transaction information and traceability information based on the settlement information, the payment information and the payment method in the transaction process, upload the transaction data to the management platform of the cloud server and upload the traceability information to the product traceability platform.

[0042] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A weighing settlement method based on a smart electronic scale, characterized by, The method comprises the following steps: When it is detected that a scale disc of a smart electronic scale is placed with a to-be-weighed settlement article, a user image and initial image information and initial weight information of the to-be-weighed settlement article are collected; The initial image information and the initial weight information are processed to obtain target image information and target weight information; Image recognition is performed on the target image information to obtain an identification result; If the identification result is that there are at least two similar candidate articles, user information and historical purchase data associated with the user image are obtained; According to the user information, the historical purchase data, the identification result and the target weight information, a target article is determined from at least two similar candidate articles; Single price information of the target article is obtained, and settlement information of the to-be-weighed settlement article is determined according to the single price information and the target weight information.

2. The method of claim 1, wherein, The step of processing the initial image information and the initial weight information to obtain target image information and target weight information comprises the following steps: The initial weight information is denoised to obtain denoised weight data; When the fluctuation error of the denoised weight data for a continuous preset number of times is less than a preset weight data fluctuation error, the denoised weight data is taken as the target weight information; Normalization operation is performed on the initial image data to obtain the target image information.

3. The method of claim 1, wherein, The step of performing image recognition on the target image information to obtain an identification result comprises the following steps: The target image information is input into a lightweight convolutional neural network model to extract article features to obtain an article feature vector; The article feature vector is compared with a preset commodity feature library in terms of cosine similarity to screen out the candidate articles with a similarity greater than a preset similarity to obtain the identification result.

4. The method of claim 1, wherein, The step of determining a target article from at least two similar candidate articles according to the user information, the historical purchase data, the identification result and the target weight information comprises the following steps: The weight weight coefficient of the candidate article is determined according to the historical purchase data, the identification result and the target weight information; The user purchase weight coefficient of the candidate article is determined according to the user information, and the purchase frequency coefficient of the candidate article is determined according to the historical purchase data; The comprehensive score of the candidate article is calculated according to the weight weight coefficient, the user purchase weight coefficient and the purchase frequency coefficient; The candidate article with the highest comprehensive score is selected as the target article.

5. The method of claim 4, wherein, The step of determining the weight weight coefficient of the candidate article according to the historical purchase data, the identification result and the target weight information comprises the following steps: The historical purchase weight of a preset time corresponding to the candidate article is obtained from the historical purchase data according to the identification result; The historical weight mean value of the candidate article is calculated according to the historical purchase weight of each candidate article; The closeness of the candidate article is calculated according to the target weight information and the historical weight mean value of each candidate article; The weight weight coefficient is given to the candidate article with the highest closeness.

6. The method of claim 4, wherein, The step of determining the user purchase weight coefficient of the candidate item according to the user information and determining the purchase frequency coefficient of the candidate item according to the historical purchase data comprises: dividing the age information in the user information into youth, middle age and old age; calculating the historical single price according to the consumption amount and the purchase times in the user information, and dividing the user consumption habit into high-end, middle-end and low-end according to the historical single price; determining the user purchase weight coefficient of the candidate item according to the age information and the user consumption habit; determining the purchase frequency coefficient according to the purchase times of the candidate item in the historical purchase data.

7. The method according to any one of claims 1 to 6, characterized in that, After the step of determining the settlement information of the to-be-weighed settlement item according to the unit price information and the target weight information, the method further comprises: displaying the settlement information and voice broadcasting the total price information in the settlement information, and displaying the payment two-dimensional code; when receiving the payment information, voice playing the payment state and the payment amount; transmitting the settlement information and the payment information to the printer for printing; generating transaction information and trace information according to the settlement information, the payment information and the payment method in the transaction process, uploading the transaction data to the management platform of the cloud server and uploading the trace information to the product trace platform.

8. A weighing settlement system based on a smart electronic scale, characterized by, The intelligent electronic scale comprises a control processing module and a weighing module, an image acquisition module and a commodity identification module connected with the control processing module, wherein the weighing module is used to obtain the initial weight information of the to-be-weighed settlement item through a high-precision sensor, and transmit the initial weight information to the control processing module for processing to obtain target weight information; the image acquisition module is used to acquire user images and initial image data of the to-be-weighed settlement item at multiple angles, and transmit the initial image data to the control processing module for processing to obtain target image information; the commodity identification module identifies the target image information to obtain an identification result, when the identification result is at least two similar candidate items, acquires user information and historical purchase data associated with the user images, determines the target item from at least two similar candidate items according to the user information, the historical purchase data, the identification result and the target weight information; the control processing module is used to determine the settlement information of the to-be-weighed settlement item according to the target item and the unit price information corresponding to the target item.

9. The system of claim 8, wherein, The intelligent electronic scale further comprises a code scanning payment module, a display module, an input module and a voice broadcast module, wherein the code scanning payment module is used to perform a payment operation; the display module is used to display settlement information; the input module is used to assist in inputting or correcting the commodity information of the to-be-weighed settlement item; and the voice broadcast module is used to broadcast the settlement information.

10. The system of claim 8, wherein, The weighing settlement system based on the intelligent electronic scale further comprises a cloud server, a printer and a traceability platform in communication connection with the intelligent electronic scale; the intelligent electronic scale further comprises a communication module, which is configured to send the settlement information to the printer for printing, and upload the transaction information and the traceability information generated by the control processing module to the cloud server and the traceability platform respectively.

Images