Intelligent settlement platform and settlement system

The intelligent settlement system uses RFID plate tags with embedded image features and camera-assisted food recognition to automate accurate payment pricing, addressing manual errors and inefficiencies in traditional catering settlement methods.

CN118823931BActive Publication Date: 2024-12-03HUNAN FUGUIOT TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411304776.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-19
Publication Date
2024-12-03
Estimated Expiration
2044-09-19

AI Technical Summary

Technical Problem

Traditional catering settlement methods rely on manual ordering and price calculation, leading to errors and inefficiencies, and existing RFID-based systems require additional manual intervention when tags differ from actual food, affecting customer experience and operational efficiency.

Method used

An intelligent settlement system using an RFID plate tag with embedded image features, processed through a camera unit to identify actual food items, determining the payment price based on actual images and tag information, utilizing image recognition and artificial intelligence for accurate food recognition.

Benefits of technology

Enables fast, error-free settlement without human intervention, improving customer satisfaction and operational efficiency by accurately determining the payment price even when incorrect food is placed, and reducing queue times during peak hours.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118823931B_ABST
    Figure CN118823931B_ABST
Patent Text Reader

Abstract

The present invention discloses an intelligent settlement counter and a settlement system, comprising: an article settlement module and an assignment module connected thereto, wherein the assignment module writes food parameters and image features into an RFID plate tag of a corresponding food and a database, relates to the technical field of commodity settlement, and solves the technical problem that when the RFID tag is different from the actual food, additional manual operation intervention is still required; by assigning values ​​to different types of plates in advance through the assignment module, the price of the food can be obtained by identifying the RFID plate tag located in the plate when in use; even if non-corresponding food is placed on the plate, an image of the food can be obtained through a camera unit when settlement is performed, and the actual payment price is determined based on the actual image obtained and the information recorded in the RFID plate tag, and settlement can be achieved in this way without manual participation throughout the process.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of commodity settlement, and in particular is an intelligent settlement platform and a settlement system. Background Art

[0002] In the catering industry, customer service experience, operational efficiency and cost control are key factors that determine the competitiveness of an enterprise. Among them, the settlement link, as an important part of the customer's dining process, directly affects customer satisfaction and the restaurant's operational efficiency.

[0003] The traditional catering settlement method mainly relies on manual ordering, price calculation and cashiering, which has many problems. First, manual operation is prone to errors, such as ordering the wrong dishes, calculating the wrong amount, etc., which not only affects the customer experience, but also causes economic losses to the restaurant. Secondly, during peak hours, manual settlement is slow, which easily causes customers to queue up, reduces the turnover rate, and affects the restaurant's operating efficiency. Although there is a system that can automatically settle through RFID tags, when the RFID tag is different from the actual food, it will cause problems with the settlement price of the food, and additional manual intervention is still required, affecting the customer's consumption experience. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes an intelligent checkout counter and a settlement system for solving the technical problem that when the RFID tag is different from the actual food, additional manual operation intervention is still required. The present invention obtains an image of the food through a camera unit during settlement, and determines the actual payment price based on the actual image obtained and the information recorded in the RFID plate tag, thereby solving the above problem.

[0005] To achieve the above-mentioned object, the first aspect of the present invention provides a settlement system for an intelligent settlement station, comprising: an item settlement module and a value assignment module connected thereto;

[0006] The assignment module: identifies the RFID plate tag in the settlement area of ​​the settlement terminal, and obtains the image features of the food corresponding to the RFID plate tag; wherein the image features include histogram standard data, a standard color value set and a standard grayscale set; and,

[0007] Write food parameters and image features into the RFID plate tag and database of the corresponding food to obtain tag information; wherein the food parameters include the unit price and the name of the food;

[0008] The item settlement module: when the items to be settled are placed in the settlement area of ​​the settlement terminal, the food in the plate is identified based on the camera unit on the settlement terminal to obtain actual information; and

[0009] The actual payment price is obtained based on the identified tag information and actual information.

[0010] In the above manner, even if inappropriate food is placed on the plate, the image of the food will be acquired through the camera unit during settlement, and the actual payment price will be determined based on the actual image acquired and the information recorded in the RFID plate tag.

[0011] The method of identifying the food in the plate based on the camera unit on the settlement terminal includes:

[0012] S1: Determine whether there is an RFID plate tag in the settlement area through the RFID sensor in the settlement terminal; if yes, control the camera unit to obtain an image of the settlement area, mark it as a captured image, and jump to S2; if no, continue to determine whether there is an RFID tag in the detection area;

[0013] S2: Remove the background technology in the captured image through image recognition technology to obtain the actual image;

[0014] S3: extract the frequency distribution of each color channel in the actual image through the color histogram algorithm and then normalize it to obtain the actual data of the histogram;

[0015] S4: randomly select several regions in the actual image, extract the RGB color values ​​and grayscale values ​​of all pixels in the several regions, and mark them as comparison features;

[0016] S5: determining a food screening range based on the comparison feature and the image feature; wherein the food screening range is a number of food names;

[0017] S6: Determine actual information in the food screening range based on the histogram standard data and the histogram actual data.

[0018] By processing the captured images in this way, more accurate actual information can be determined.

[0019] Preferably, the step of obtaining the image features of the food corresponding to the RFID plate tag includes:

[0020] Acquire food images corresponding to RFID plate labels respectively, remove backgrounds from the food images according to image recognition technology, and obtain screening images;

[0021] The frequency distribution of each color channel in the filtered image is extracted by the color histogram algorithm and then normalized to obtain the histogram standard data;

[0022] Extract the RGB color value and grayscale value of all pixels in the screened image; perform value expansion processing on the RGB color value and grayscale value of all pixels, and integrate them into a standard color value set and a standard grayscale set respectively.

[0023] It should be noted that food images can be obtained by downloading images of the food or by photographing the food.

[0024] Preferably, the step of expanding the RGB color values ​​and grayscale values ​​of all pixels includes:

[0025] Extract the RGB color value and grayscale value of all pixels;

[0026] The values ​​corresponding to the R, G, and B values ​​of the RGB color values ​​of all pixels [R, G, B] i Adjust to [R±n, G±n, B±n] i , and obtain several standard color value subsets; where i and n are positive integers;

[0027] The gray value level Gs of all pixels i Adjust to [Gs±n] i , and obtain several standard grayscale subsets.

[0028] It should be noted that the value of n can be determined according to the number of different RGB color values ​​in a number of regions, and i represents the i-th pixel among all pixels.

[0029] By performing value expansion processing on the RGB color value and the grayscale value, it is possible to avoid the situation where no matching result occurs due to the color difference of the food when subsequently matching the standard color value set and the standard grayscale set with the comparison feature.

[0030] Preferably, the step of writing food parameters and image features into the RFID plate tag and database of the corresponding food includes:

[0031] Obtain the food name and food unit price corresponding to the RFID plate tag, and write the food name, food unit price and image features corresponding to the RFID plate tag into the corresponding RFID plate tag;

[0032] A mapping relationship between food names, food unit prices and image features is established, and the food names, food unit prices, image features and mapping relationships corresponding to several RFID plate tags are written into the database in turn.

[0033] Preferably, the determining of the food screening range based on the comparison feature and the image feature includes:

[0034] Extract the RGB color values ​​of all pixels and the grayscale values ​​of all pixels in the comparison feature;

[0035] Match the RGB color values ​​of all pixels in the comparison feature with several standard color value sets in the database respectively to obtain several standard color value sets containing all the RGB color values, and mark the food types corresponding to the several standard color value sets as the preliminary screening range;

[0036] The grayscale values ​​of all pixels in the comparison feature are matched with several standard grayscale sets corresponding to the preliminary screening range to obtain several standard grayscale sets containing the grayscale values, and the food types corresponding to the several standard grayscale sets are marked as the food screening range.

[0037] Preferably, the determining of actual information in the food screening range based on the histogram standard data and the histogram actual data comprises:

[0038] Extracting actual data of the histogram and several standard data of the histogram corresponding to the food types in the food screening range;

[0039] By respectively calculating the similarity between a number of histogram standard data and the histogram actual data, a number of similar data are obtained; the number of similar data are arranged, and based on the arrangement result, the histogram actual data with the greatest similarity is obtained, and the food parameters corresponding to the histogram actual data with the greatest similarity are marked as actual information.

[0040] Preferably, obtaining the actual payment price according to the identified label information and actual information includes:

[0041] Determine whether the actual food is consistent with the food type on the RFID plate tag;

[0042] Yes, the actual payment price is obtained based on the unit price of the food in the RFID plate tag;

[0043] No, the food parameters and image features in the RFID plate label as well as the comparative features and histogram actual data of the actual image are input into the food recognition module to obtain the identified food; the corresponding food unit price in the database is matched according to the identified food and the actual payment price is obtained; wherein, the food recognition module is established based on the artificial intelligence model.

[0044] The food recognition module can avoid confusion when identifying foods of different types but similar images, further improving the accuracy of recognition.

[0045] Preferably, the food identification module is established based on an artificial intelligence model, including:

[0046] Obtain food parameters and image features in the RFID plate tag, comparative features of the actual image and histogram actual data, and identify the food in the training library;

[0047] Integrate food parameters and image features in RFID plate tags, comparative features of actual images and histogram actual data, and identified foods into several sets of training data and several sets of test data;

[0048] The artificial intelligence model is trained through several groups of experimental data; the trained artificial intelligence model is tested using the training data, and the artificial intelligence model is adjusted according to the test results; finally, the input is the food parameters and image features in the RFID plate label, the comparative features of the actual image and the actual data of the histogram, and the output is a food recognition model for identifying food.

[0049] The second aspect of the present invention provides an intelligent settlement station for implementing a settlement system of an intelligent settlement station in the first aspect, comprising a base station, a workbench is arranged above the base station, and a display module and a camera module are arranged on one side of the base station;

[0050] The display module is used to display food parameters on the workbench and to control the settlement system, and the camera module is used to collect images on the workbench and to control it through the camera unit.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. In the present invention, different types of plates are assigned values ​​in advance by the assignment module. When in use, the price of food can be obtained by identifying the RFID plate tag located in the plate. Even if non-corresponding food is placed on the plate, the image of the food will be obtained by the camera unit during settlement, and the actual payment price will be determined based on the actual image obtained and the information recorded in the RFID plate tag. In this way, settlement can be achieved without human intervention throughout the process, so as to improve the settlement speed of customers and the operating efficiency of manufacturers.

[0053] 2. In the present invention, when assigning values ​​to RFID plate tags, the image features of the corresponding food are assigned to the database and the RFID plate tags together. If non-corresponding food is placed on the plate later, by obtaining the comparison features and the actual data of the histogram in the actual image, it can be compared with the image features in the database, so that the name of the food on the plate and the unit price of the food can be accurately determined, and then the actual payment price can be accurately obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 A schematic diagram of a flow chart of a settlement system of an intelligent settlement station of the present invention;

[0056] Figure 2 It is a logic block diagram for obtaining actual information in the present invention;

[0057] Figure 3 A logic block diagram for obtaining the actual payment price in the present invention;

[0058] Figure 4 It is a schematic diagram of the external structure of the intelligent settlement station in the present invention.

[0059] In the figure: 1. base; 2. workbench; 3. display module; 4. camera module. DETAILED DESCRIPTION

[0060] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] See also Figure 1-Figure 4 , the first aspect of the present invention provides an intelligent settlement station and a settlement system, including: an item settlement module and a value assignment module connected thereto;

[0062] The assignment module: identifies the RFID plate tag in the settlement area of ​​the settlement terminal, and obtains the image features of the food corresponding to the RFID plate tag; wherein the image features include histogram standard data, a standard color value set and a standard grayscale set; and,

[0063] Write food parameters and image features into the RFID plate tag and database of the corresponding food to obtain tag information; wherein the food parameters include the unit price and the name of the food;

[0064] The item settlement module: when the items to be settled are placed in the settlement area of ​​the settlement terminal, the food in the plate is identified based on the camera unit on the settlement terminal to obtain actual information; and

[0065] The actual payment price is obtained based on the identified tag information and actual information.

[0066] In this embodiment, obtaining the image features of the food corresponding to the RFID plate tag includes:

[0067] By taking photos of each food, the food images corresponding to the RFID plate tags are obtained respectively, and the background in the food images is removed according to the image recognition technology, and the food itself is focused to obtain the screening images;

[0068] The frequency distribution of each color channel in the filtered image is extracted by the color histogram algorithm and then normalized to obtain the histogram standard data;

[0069] Extract the RGB color value and grayscale value of all pixels in the screened image; perform value expansion processing on the RGB color value and grayscale value of all pixels, and integrate them into a standard color value set and a standard grayscale set respectively.

[0070] Furthermore, the RGB color values ​​and grayscale values ​​of all pixels are expanded, including:

[0071] Extract the RGB color value and grayscale value of all pixels;

[0072] The values ​​corresponding to the R, G, and B values ​​of the RGB color values ​​of all pixels [R, G, B] i Adjust to [R±n, G±n, B±n] i , and obtain several standard color value subsets; where i and n are positive integers;

[0073] The gray value level Gs of all pixels i Adjust to [Gs±n] i , and obtain several standard grayscale subsets.

[0074] By performing value expansion processing on the RGB color value and the grayscale value, it is possible to avoid the situation where no matching result occurs due to the color difference of the food when subsequently matching the standard color value set and the standard grayscale set with the comparison feature.

[0075] For example, the numerical values ​​corresponding to the R value, G value, and B value of the RGB color value in the pixel are [R100, G100, B100]. When n is 1, the standard color value subsets corresponding to [R100, G100, B100] are [R100, G100, B100], [R101, G100, B100], [R100, G101, B100], [R100, G100, B101], [R101, G101, B100], [R 101,G100,B101], [R100,G101,B101], [R101,G101,B101], [R99,G100,B100], [R100,G99,B100], [R100,G100,B99], [R99,G99,B100], [R100,G99,B99], [R99,G100,B99] and [R99,G99,B99]; similarly, the grayscale value level Gsi is 100, then the corresponding standard grayscale subsets are 99, 100 and 101.

[0076] In this embodiment, the food parameters and image features are written into the RFID plate tag and database of the corresponding food, including:

[0077] Obtain the food name and food unit price corresponding to the RFID plate tag, and write the food name, food unit price and image features corresponding to the RFID plate tag into the corresponding RFID plate tag;

[0078] A mapping relationship between food names, food unit prices and image features is established, and the food names, food unit prices, image features and mapping relationships corresponding to several RFID plate tags are written into the database in turn.

[0079] In this embodiment, in order to improve the accuracy of the actual information, the food in the plate is identified based on the camera unit on the settlement terminal, including:

[0080] S1: Determine whether there is an RFID plate tag in the settlement area through the RFID sensor in the settlement terminal; if yes, control the camera unit to obtain an image of the settlement area, mark it as a captured image, and jump to S2; if no, continue to determine whether there is an RFID tag in the detection area;

[0081] S2: Remove the background technology in the captured image through image recognition technology to obtain the actual image;

[0082] S3: extract the frequency distribution of each color channel in the actual image through the color histogram algorithm and then normalize it to obtain the actual data of the histogram;

[0083] S4: randomly select several regions in the actual image, extract the RGB color values ​​and grayscale values ​​of all pixels in the several regions, and mark them as comparison features;

[0084] S5: determining a food screening range based on the comparison feature and the image feature; wherein the food screening range is a number of food names;

[0085] S6: Determine actual information in the food screening range based on the histogram standard data and the histogram actual data.

[0086] Furthermore, the food screening range is determined based on the comparison features and image features, including:

[0087] Extract the RGB color values ​​of all pixels and the grayscale values ​​of all pixels in the comparison feature;

[0088] Match the RGB color values ​​of all pixels in the comparison feature with several standard color value sets in the database respectively to obtain several standard color value sets containing all the RGB color values, and mark the food types corresponding to the several standard color value sets as the preliminary screening range;

[0089] The grayscale values ​​of all pixels in the comparison feature are matched with several standard grayscale sets corresponding to the preliminary screening range to obtain several standard grayscale sets containing the grayscale values, and the food types corresponding to the several standard grayscale sets are marked as the food screening range.

[0090] For example: if the RGB values ​​of a pixel are (200, 150, 100) or (100, 100, 100), the system will search the database for all food standard color value sets that contain or are close to these two RGB color values.

[0091] This RGB value matches in the standard color value sets A, B, and C, indicating that the food product is likely to be type A, B, or C.

[0092] In order to determine the closest food type in the food screening range, actual information is determined in the food screening range based on the histogram standard data and the histogram actual data, including:

[0093] Extracting actual data of the histogram and several standard data of the histogram corresponding to the food types in the food screening range;

[0094] By respectively calculating the similarity between a number of histogram standard data and the histogram actual data, a number of similar data are obtained; the number of similar data are arranged, and based on the arrangement result, the histogram actual data with the greatest similarity is obtained, and the food parameters corresponding to the histogram actual data with the greatest similarity are marked as actual information.

[0095] It should be noted that when calculating the similarity between the histogram standard data and the histogram actual data, the similarity is measured by calculating the size of the intersection of the two color histograms. It is mainly to calculate the sum of the minimum values ​​of the corresponding elements in the two histogram vectors; the larger the intersection, the higher the similarity. Specifically, for each color channel in the two color histograms, the intersection of the corresponding color values ​​is calculated, that is, the sum of the minimum values ​​of the number of occurrences of the color value in the two histograms.

[0096] As the last step of this embodiment, the actual payment price is obtained according to the identified tag information and the actual information, including:

[0097] Determine whether the actual food is consistent with the food type on the RFID plate tag;

[0098] Yes, the actual payment price is obtained based on the unit price of the food in the RFID plate tag;

[0099] No, the food parameters and image features in the RFID plate label as well as the comparative features and histogram actual data of the actual image are input into the food recognition module to obtain the identified food; the corresponding food unit price in the database is matched according to the identified food and the actual payment price is obtained; wherein, the food recognition module is established based on the artificial intelligence model.

[0100] The food recognition module can avoid confusion when identifying foods of different types but similar images, further improving the accuracy of recognition.

[0101] Furthermore, the food recognition module is established based on the artificial intelligence model, including:

[0102] Obtain food parameters and image features in the RFID plate tag, comparative features of the actual image and histogram actual data, and identify the food in the training library;

[0103] Integrate food parameters and image features in RFID plate tags, comparative features of actual images and histogram actual data, and identified foods into several sets of training data and several sets of test data;

[0104] The artificial intelligence model is trained through several groups of experimental data; the trained artificial intelligence model is tested using the training data, and the artificial intelligence model is adjusted according to the test results; finally, the input is the food parameters and image features in the RFID plate label, the comparative features of the actual image and the actual data of the histogram, and the output is a food recognition model for identifying food.

[0105] As a further embodiment of the present invention, when the actual food is inconsistent with the food type in the RFID plate label, the group of related data can be recorded and manually judged through the background, and the judged data can be used as data in the training library for re-training and verification.

[0106] As a further embodiment of the present invention, the system further includes a settlement module, which is used to charge diners a corresponding amount of fees based on the unit price of the food.

[0107] An intelligent settlement station, used to implement a settlement system of an intelligent settlement station of the first aspect, comprising a base station 1, a workbench 2 is arranged above the base station 1, and a display module 3 and a camera module 4 are arranged on one side of the base station 1;

[0108] Specifically, the workbench 2 is divided into a waiting area and a recognition area, and the recognition area is located at one side of the camera module 4. After the recognition is completed, the food on the waiting area can be put into the recognition area.

[0109] The display module 3 is used to display food parameters in the waiting area on the workbench 2 and to control the settlement system. The camera module 4 is used to collect images on the workbench 2 and to control the camera unit to realize the photography and recognition of dishes in the waiting area.

[0110] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0111] The working principle of the present invention is as follows: an RFID plate tag in a settlement area in a settlement terminal is identified, image features of food corresponding to the RFID plate tag are obtained, food parameters and image features are written into the RFID plate tag of the corresponding food and into a database to obtain tag information; when items to be settled are placed in the settlement area of ​​the settlement terminal, the RFID sensor in the settlement terminal is used to determine whether an RFID plate tag exists in the settlement area; if yes, the camera unit is controlled to obtain an image of the settlement area, and the image is marked as a photographed image, and the process is skipped to S2; if no, the detection area is continuously determined to determine whether an RFID tag exists, the background technology in the photographed image is eliminated by image recognition technology to obtain an actual image, the frequency distribution of each color channel in the actual image is extracted by a color histogram algorithm, and then normalized to obtain histogram actual data, several areas are randomly selected in the actual image, RGB color values ​​and gray values ​​of all pixels in the several areas are extracted, and the values ​​are marked as comparison features, and a food screening range is determined based on the comparison features and image features; wherein the food screening range is several food names; actual information is determined in the food screening range based on the histogram standard data and the histogram actual data; the actual payment price is obtained according to the identified tag information and actual information.

[0112] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A settlement system for an intelligent settlement counter, characterized in that: include: Item settlement module and the value assignment module connected thereto; The assignment module: identifies the RFID plate tag in the settlement area of ​​the settlement terminal, and obtains the image features of the food corresponding to the RFID plate tag; wherein the image features include histogram standard data, a standard color value set and a standard grayscale set; and, Write food parameters and image features into the RFID plate tag and database of the corresponding food to obtain tag information; wherein the food parameters include the unit price and the name of the food; The item settlement module: when the items to be settled are placed in the settlement area of ​​the settlement terminal, the food in the plate is identified based on the camera unit on the settlement terminal to obtain actual information; and Obtaining the actual payment price based on the identified label information and actual information; The method of identifying the food in the plate based on the camera unit on the settlement terminal includes: S1: Determine whether there is an RFID plate tag in the settlement area through the RFID sensor in the settlement terminal; if yes, control the camera unit to obtain an image of the settlement area, mark it as a captured image, and jump to S2; if no, continue to determine whether there is an RFID tag in the detection area; S2: Remove the background technology in the captured image through image recognition technology to obtain the actual image; S3: extract the frequency distribution of each color channel in the actual image through the color histogram algorithm and then normalize it to obtain the actual data of the histogram; S4: randomly select several regions in the actual image, extract the RGB color values ​​and grayscale values ​​of all pixels in the several regions, and mark them as comparison features; S5: determining a food screening range based on the comparison feature and the image feature; wherein the food screening range is a number of food names; S6: determining actual information in the food screening range based on the histogram standard data and the histogram actual data; The method of determining the food screening range based on the comparison feature and the image feature includes: Extract the RGB color values ​​of all pixels and the grayscale values ​​of all pixels in the comparison feature; Match the RGB color values ​​of all pixels in the comparison feature with several standard color value sets in the database respectively to obtain several standard color value sets containing all the RGB color values, and mark the food types corresponding to the several standard color value sets as the preliminary screening range; The grayscale values ​​of all pixels in the comparison feature are matched with several standard grayscale sets corresponding to the preliminary screening range to obtain several standard grayscale sets containing the grayscale values, and the food types corresponding to the several standard grayscale sets are marked as the food screening range.

2. The settlement system of an intelligent settlement counter according to claim 1, characterized in that: The step of obtaining the image features of the food corresponding to the RFID plate tag includes: Acquire food images corresponding to RFID plate labels respectively, remove backgrounds from the food images according to image recognition technology, and obtain screening images; The frequency distribution of each color channel in the filtered image is extracted by the color histogram algorithm and then normalized to obtain the histogram standard data; Extract the RGB color value and grayscale value of all pixels in the screened image; perform value expansion processing on the RGB color value and grayscale value of all pixels, and integrate them into a standard color value set and a standard grayscale set respectively.

3. The settlement system of an intelligent settlement counter according to claim 2, characterized in that: The step of expanding the RGB color values ​​and grayscale values ​​of all pixels includes: Extract the RGB color value and grayscale value of all pixels; The values ​​corresponding to the R, G, and B values ​​of the RGB color values ​​of all pixels [R, G, B] i Adjust to [R±n, G±n, B±n] i , and obtain several standard color value subsets; where i and n are positive integers; The gray value level Gs of all pixels i Adjust to [Gs±n] i , and obtain several standard grayscale subsets.

4. The settlement system of an intelligent settlement counter according to claim 1, characterized in that: The step of writing food parameters and image features into the RFID plate tag and database of the corresponding food includes: Obtain the food name and food unit price corresponding to the RFID plate tag, and write the food name and food unit price corresponding to the RFID plate tag into the corresponding RFID plate tag; A mapping relationship between food names, food unit prices and image features is established, and the food names, food unit prices, image features and mapping relationships corresponding to several RFID plate tags are written into the database in turn.

5. The settlement system of an intelligent settlement counter according to claim 1, characterized in that: The actual information is determined in the food screening range based on the histogram standard data and the histogram actual data, including: Extracting actual data of the histogram and several standard data of the histogram corresponding to the food types in the food screening range; By respectively calculating the similarity between a number of histogram standard data and the histogram actual data, a number of similar data are obtained; the number of similar data are arranged, and based on the arrangement result, the histogram actual data with the greatest similarity is obtained, and the food parameters corresponding to the histogram actual data with the greatest similarity are marked as actual information.

6. The settlement system of an intelligent settlement counter according to claim 1, characterized in that: The obtaining of the actual payment price according to the identified label information and the actual information includes: Determine whether the actual food is consistent with the food type on the RFID plate tag; Yes, the actual payment price is obtained based on the unit price of the food in the RFID plate tag; No, the food parameters and image features in the RFID plate label as well as the comparative features and histogram actual data of the actual image are input into the food recognition module to obtain the identified food; the corresponding food unit price in the database is matched according to the identified food and the actual payment price is obtained; wherein, the food recognition module is established based on the artificial intelligence model.

7. The settlement system of an intelligent settlement counter according to claim 6, characterized in that: The food recognition module is established based on an artificial intelligence model and includes: Obtain food parameters and image features in the RFID plate tag, comparative features of the actual image and histogram actual data, and identify the food in the training library; Integrate food parameters and image features in RFID plate tags, comparative features of actual images and histogram actual data, and identified foods into several sets of training data and several sets of test data; The artificial intelligence model is trained through several groups of experimental data; the trained artificial intelligence model is tested using the training data, and the artificial intelligence model is adjusted according to the test results; finally, the input is the food parameters and image features in the RFID plate label, the comparative features of the actual image and the actual data of the histogram, and the output is a food recognition model for identifying food.

8. An intelligent settlement station, used to implement the settlement system of an intelligent settlement station as claimed in any one of claims 1 to 7, characterized in that: It comprises a base platform (1), a workbench (2) is arranged above the base platform (1), and a display module (3) and a camera module (4) are arranged on one side of the base platform (1); The display module (3) is used to display food parameters on the workbench (2) and to control the settlement system, and the camera module (4) is used to capture images on the workbench (2).