AI-based intelligent selection system and method for customers to purchase dry goods products

Abstract

The present invention relates to the technical field of dry goods purchase and selection, and discloses an AI-based intelligent selection system and method for customers to purchase dry goods products. The system includes a dry goods product selection search module, a dry goods product selection processing module, and a dry goods product selection feedback module; based on the framed selection image information of a single dry goods object at the dry goods product selection site, an intelligent search algorithm and a qualified overall feature image library of a target dry goods product are combined to perform a comprehensive intelligent evaluation of the selection quality of a single dry goods object at the dry goods product selection site, thereby achieving efficient and accurate selection of dry goods products; the selection result image of a single dry goods object at the dry goods product selection site is combined with image feature analysis and the framed selection image of the dry goods object at the dry goods product selection site for image matching and combination to scientifically construct the dry goods product selection site result image information, thereby achieving accurate marking of the selection quality of a single dry goods object at the dry goods product selection site, thereby improving the customer's dry goods purchasing experience and dry goods purchasing quality.

Description

Technical Field

[0001] The present invention relates to the technical field of dry goods purchase and selection, and specifically to an AI-based intelligent selection system and method for customers to purchase dry goods products. Background Art

[0002] Dried goods generally refer to condiments and foods that have had their moisture removed by air-drying or sun-drying. Common dried goods include wood ear mushrooms, seaweed, shiitake mushrooms, red dates, cinnamon, chili peppers, Sichuan peppercorns, aniseed, cumin, peppercorns, goji berries, longans, peanuts, tangerine peel, and raisins. Dry goods are categorized by purpose into dried fruits, dried vegetables, seasonings, grains, beverages, medicinal foods, and preserved fruits. The dry goods selection process is highly demanding on the customer's experience, as factors such as color, shape, smell, origin, shelf life, and brand must be considered. Choosing the right quality dry goods from a wide variety of products is a crucial step in purchasing dry goods. Existing product recommendation systems cannot intelligently select high-quality dry goods for customers on-site based on their variety, origin, or brand, nor can they intuitively present the selection results based on image labels, reducing the customer's dry goods shopping experience.

[0003] A Chinese invention patent application with publication number CN115526674A discloses a product recommendation method, storage medium, and product, which obtain a user's product demand information through a user terminal, determine recommended products based on the product demand information, and send the recommended products to the user terminal for display. The user can quickly select the products to be purchased from the recommended products according to his or her needs; however, the above technical solution cannot select high-quality products for the user at the shopping site according to user needs, nor can it provide intuitive feedback on the distinction between high-quality products and problematic products on the site. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In order to solve the problem that the above-mentioned existing product recommendation system cannot intelligently select high-quality dry goods products for customers on-site based on the variety, origin and brand of dry goods products, and cannot intuitively present the dry goods product selection results based on image identification, which reduces the customer's dry goods product shopping experience, the above purposes are achieved: online collection of dry goods product feature information and dry goods product selection site images, accurate search of target dry goods product single dry goods object frame selection calibration images, intelligent matching of target dry goods product qualified overall feature image library, scientific generation of dry goods object frame selection images on dry goods product selection site, autonomous segmentation of dry goods object frame selection images on dry goods product selection site, intelligent analysis of dry goods product selection site single dry goods object qualified results, scientific matching of dry goods product selection site single dry goods object qualified analysis result identification watermark image, accurate generation of dry goods product selection site single dry goods object selection result image, scientific construction of dry goods product selection site selection result image, and visually differentiated feedback of dry goods product on-site selection results.

[0006] (2) Technical solution

[0007] The present invention is implemented through the following technical solution: an AI-based intelligent selection method for customers to purchase dry goods products, characterized in that the method includes the following steps:

[0008] S1. Collecting target dry goods product feature text data and dry goods product selection site image data;

[0009] S2. Performing a search process for a single dry goods object frame selection calibration reference image in the target dry goods product selection scene image based on the target dry goods product feature text data and the dry goods product single dry goods object frame selection calibration image data to generate the target dry goods product single dry goods object frame selection calibration image data;

[0010] S3. Performing matching processing on the appearance feature image library of qualified dry goods products required for target dry goods product selection and analysis based on the target dry goods product feature text data and the qualified overall feature image library of dry goods products to generate a qualified overall feature image library of target dry goods products;

[0011] S4. Performing a framing and calibration process on a single dry goods object in the target dry goods product selection scene image based on the dry goods product selection scene image data and the framing and calibration image data of a single dry goods object of the target dry goods product to generate framing and calibration image data of the dry goods product selection scene dry goods object, and performing a framing and segmentation process on the single dry goods object image of the dry goods product selection scene dry goods object to generate framing and calibration image data of the single dry goods object of the dry goods product selection scene dry goods object;

[0012] S5. Analyze and process the qualified results of the selection of the individual dry goods object at the dry goods product selection site based on the frame selection image data of the individual dry goods object at the dry goods product selection site and the qualified overall feature image library of the target dry goods product, to generate qualified result analysis data of the individual dry goods object at the dry goods product selection site;

[0013] S6. Perform identification watermark image matching processing required for different selection quality qualified analysis results of individual dry goods objects at the dry goods product selection site based on the qualified result analysis data of the individual dry goods objects at the dry goods product selection site and the identification watermark image data of the qualified analysis results of the dry goods objects, generate the identification watermark image data of the qualified analysis results of the individual dry goods objects at the dry goods product selection site, and perform selection result image identification processing of the individual dry goods objects at the dry goods product selection site on the frame selection image data of the individual dry goods objects at the dry goods product selection site, generate the selection result image data of the individual dry goods objects at the dry goods product selection site;

[0014] S7. Construct image data of the on-site results of the dry goods product selection and perform a dry goods product on-site selection result feedback operation.

[0015] Preferably, the steps of collecting target dry goods product feature text data and dry goods product selection site image data are as follows:

[0016] S11. Input the characteristic text information of the dry goods product to be purchased by the customer online through the mobile terminal, and generate the characteristic text data of the target dry goods product The target dry food product characteristic text data includes the name text data, variety text data, origin text data and brand text data of the dry food product; the mobile terminal includes any one of a smart phone, a smart watch and a tablet computer;

[0017] Use mobile terminals to capture on-site image information of dry goods products that customers are waiting to purchase in an unpacked state, and generate on-site image data of dry goods product selection .

[0018] Preferably, the steps of searching for a reference image for the frame selection and calibration of a single dry goods object in the target dry goods product selection scene image based on the target dry goods product feature text data and the frame selection and calibration image data of a single dry goods object of the dry goods product are as follows:

[0019] S21. Create a single dry product object selection and calibration image data set , ;in Indicates the The image data of a single dry product object of the dry product corresponding to the dry product combination feature type is selected and calibrated. Indicates the maximum number of dry food product feature types; the dry food product combination feature type represents an index identification type formed by a combination of name, variety, origin, and brand feature information for searching for a specific dry food product single dry food object selection calibration image; the dry food product single dry food object selection calibration image data represents a reference image for identifying and selecting a single dry food object image in a selection scene image of different types of dry food products;

[0020] S22, the target dry goods product feature text data A set of image data for selecting and calibrating a single dry product object of the dry product Select and calibrate image data for a single dry product object in the dry product Perform dry goods product feature character matching to search for the target dry goods product feature text data The corresponding single dry goods object of the dry goods product is selected and calibrated image data , and generate the target dry goods product single dry goods object selection calibration image data through data identification , execute to generate the target dry goods product single dry goods object selection calibration image data The specific steps are as follows:

[0021] S221, initialization, update the maximum number of iterations T and update the calibration image to search for the pelican population location. The formula for updating the calibration image to search for the pelican population location is as follows: ,in Indicates the The calibrated image searches for the pelican individual in The position of the dimension, that is, The calibration images search for pelican individuals in the dimension The dry goods product single dry goods object frame selection calibration image data set The position in the search space, Indicates a random integer for position adjustment, rand indicates a random number in the range [0,1], and Indicates in The upper and lower boundaries of the problem are solved in the dimension The dry goods product single dry goods object frame selection calibration image data set Search the search space to find the text data related to the target dry goods product characteristics Matching the single dry product object of the dry product to select the calibration image data The number of upper and lower boundaries;

[0022] S222, exploration phase, calibration image search pelican to determine the location of the prey, that is, to select the calibration image data set in the dry goods product single dry goods object Search the search space to find the text data related to the target dry goods product characteristics Matching the single dry product object of the dry product to select the calibration image data The location of the calibration image is then searched for the pelican individual to the location where the dry goods product single dry goods object is located to select the calibration image data The area moves, and the calibration image search pelican adopts the approximate prey strategy to model, so that the algorithm selects the calibration image data set for the single dry goods object of the dry goods product The search space is scanned, and the target dry goods product single dry goods object is selected and calibrated to the image data The location of the prey in the dry product single dry object selection calibration image data set The search space is randomly generated, and the approximate prey strategy formula is as follows: ,in Indicates the exploration phase after the update The calibrated image searches for the pelican individual in The position of the dimension, that is, the position of the first dimension after the exploration phase update The calibration images search for pelican individuals in the dimension The dry goods product single dry goods object frame selection calibration image data set Search the search space to find the text data related to the target dry goods product characteristics Matching the single dry product object of the dry product to select the calibration image data location, For prey in Dimensional position, that is, the target dry product feature text data Matching the single dry product object of the dry product to select the calibration image data The prey is in the dimension The dry goods product single dry goods object frame selection calibration image data set The position in the search space, is the objective function value of the prey, For the The objective function value of the individual pelican is searched from the calibration images;

[0023] S223, development stage, calibration image search Pelican selects a calibration image data set for a single dry goods object of the dry goods product All the dry goods products in the search space are marked with single dry goods object frame selection image data Perform target prey hunting search, that is, select and calibrate the image data set of a single dry product object in the dry product product Search the search space to find the text data related to the target dry goods product characteristics Matching the single dry product object of the dry product to select the calibration image data , the behavior process of the calibration image search pelican is modeled. The algorithm checks the position near the calibration image search pelican position so that the algorithm converges to a better position. The calculation formula of the calibration image search pelican hunting behavior is as follows: ,in Indicates the update after the development phase The calibrated image search pelican The position of the dimension, that is, the first The calibrated images search Pelican in dimension The dry goods product single dry goods object frame selection calibration image data set Search the search space to find the text data related to the target dry goods product characteristics Matching the single dry product object of the dry product to select the calibration image data location, is a random integer of 0 or 2; t is the current number of iterations; T is the maximum number of iterations;

[0024] S224: When the algorithm meets the maximum number of iterations, output the target dry product feature text data Matching the single dry product object of the dry product to select the calibration image data ;

[0025] S225: Mark the image data of the dry goods product output in step S224 by selecting the single dry goods object Generate target dry goods product single dry goods object selection calibration image data after data identification .

[0026] Preferably, the steps of performing matching processing on the appearance feature image library of qualified dry goods products required for target dry goods product selection analysis based on the target dry goods product feature text data and the dry goods product qualified overall feature image library to generate the target dry goods product qualified overall feature image library are as follows:

[0027] S31. Establish a collection of qualified overall feature image libraries for dry goods products ,in Indicates the A qualified overall feature image library of dry goods products corresponding to the combination feature types of dry goods products; , ;in Indicates the qualified overall feature image library of the dry goods product Middle A piece of qualified overall characteristic image data of dry goods products, Indicates the qualified overall feature image library of the dry goods product the maximum number of qualified overall characteristic images of dry food products; the qualified overall characteristic image data of dry food products represents the overall appearance image information of qualified dry food products established by different types of dry food product standards;

[0028] S32, using the KD tree nearest neighbor search algorithm to A collection of image libraries with qualified overall features of the dry goods products The image library of qualified overall features of dry goods products described in Perform keyword matching on dry goods product features and search for the target dry goods product feature text data Corresponding qualified overall feature image library of the dry goods product , and build a qualified overall feature image library of target dry goods products ,in Indicates the A piece of qualified overall characteristic image data of the target dry goods product.

[0029] Preferably, based on the dry goods product selection scene image data and the target dry goods product single dry goods object frame selection calibration image data, single dry goods object frame selection calibration processing is performed in the target dry goods product selection scene image to generate dry goods product selection scene dry goods object frame selection image data and perform single dry goods object image segmentation processing on the dry goods product selection scene dry goods object frame selection image. The operation steps for generating the dry goods product selection scene single dry goods object frame selection image data are as follows:

[0030] S41, using the FLANN algorithm to select and calibrate image data based on a single dry goods object of the target dry goods product On-site image data of dry goods product selection The single dry goods object is identified in the corresponding dry goods product selection scene image, and the identified single dry goods object is calibrated using closed line frame selection in the dry goods product selection scene image, and the dry goods object frame selection image data of the dry goods product selection scene is generated. ;

[0031] S42: Frame and select image data of the dry goods object at the dry goods product selection site The framed image of a single dry goods object is segmented and extracted along the closed line of the frame calibration, and a single dry goods object framed image data set of the dry goods product selection site is generated. , ;in Indicates the image data of the dry goods object selection at the dry goods product selection site The middle segment The image data of a single dry goods object selected at the dry goods product selection site corresponding to each dry goods object, Indicates the image data of the dry goods object selection at the dry goods product selection site The maximum number of dry goods objects.

[0032] Preferably, the steps for performing analysis and processing of the qualified result of selection of a single dry goods object at the dry goods product selection site based on the frame selection image data of the single dry goods object at the dry goods product selection site and the qualified overall feature image library of the target dry goods product are as follows to generate the qualified result analysis data of the single dry goods object at the dry goods product selection site:

[0033] S51, using the FLANN algorithm to select the image data set of a single dry goods object at the dry goods product selection site Image data of a single dry goods object selected at the dry goods product selection site The image library of the target dry goods products is ordered by the number of dry goods objects and has the qualified overall characteristics. Qualified overall characteristic image data of the target dry goods product described in Perform image feature matching and generate a data set of qualified results analysis of a single dry goods object at the dry goods product selection site based on the image feature matching results ,in Indicates the image data of the dry goods object selection at the dry goods product selection site The middle segment Analytical data on the qualified results of individual dry goods objects at the selection site for dry goods products corresponding to each dry goods object;

[0034] when and The image feature matching is successful, indicating that The quality of each dry goods product is qualified and meets the selection criteria; then the qualified result analysis data of each dry goods object at the dry goods product selection site is output. To be qualified;

[0035] when and The image features were not matched successfully, indicating that If the quality of a dry goods product fails to meet the selection criteria, the qualified result analysis data of a single dry goods object at the dry goods product selection site will be output. Is unqualified.

[0036] Preferably, identification watermark image matching processing required for different selection quality qualified analysis results of individual dry goods objects at the dry goods product selection site is performed based on the qualified result analysis data of the individual dry goods object at the dry goods product selection site and the identification watermark image data of the qualified analysis result of the dry goods object, to generate the identification watermark image data of the qualified analysis result of the individual dry goods object at the dry goods product selection site, and perform selection result image identification processing of the individual dry goods object at the dry goods product selection site with the frame selection image data of the individual dry goods object at the dry goods product selection site, to generate the selection result image data of the individual dry goods object at the dry goods product selection site as follows:

[0037] S61. Establishing a watermark image data set for qualified analysis results of dry goods objects ,in Indicates the qualified result analysis data of a single dry goods object at the dry goods product selection site The watermark image data of the qualified analysis result of the corresponding dry goods object is qualified. At this time, the watermark image of the qualified analysis result of the dry goods object is a watermark image with the word "qualified"; Indicates the qualified result analysis data of a single dry goods object at the dry goods product selection site The watermark image data of the qualified analysis result of the corresponding dry goods object when it is unqualified, in this case, the watermark image of the qualified analysis result of the dry goods object is a watermark image with the word "unqualified";

[0038] S62: using a bidirectional search algorithm to analyze the qualified result data set of a single dry goods object at the dry goods product selection site Analytical data on the qualified results of individual dry goods objects at the dry goods product selection site A set of watermarked image data is numbered in order according to the quantity of the dry goods objects and marked with the qualified analysis results of the dry goods objects. The qualified analysis result watermark image data of the dry goods object is matched with the keyword of the qualified analysis result of the single dry goods object at the dry goods product selection site, and the qualified analysis data of the single dry goods object at the dry goods product selection site is searched out. The corresponding dry goods object qualified analysis result identification watermark image data, and generate a dry goods product selection site single dry goods object qualified analysis result identification watermark image data set ,in Indicates the image data of the dry goods object selection at the dry goods product selection site The middle segment Watermark image data of qualified analysis results of a single dry goods object at the dry goods product selection site corresponding to each dry goods object;

[0039] S63: a watermark image data set containing qualified analysis results of individual dry goods objects at the dry goods product selection site Watermark image data of qualified analysis results of a single dry goods object at the dry goods product selection site Image data set of single dry goods object selection based on the dry goods object quantity number and the dry goods product selection scene Image data of a single dry goods object selected at the dry goods product selection site Combine images and generate a dataset of image data of individual dry goods object selection results at the dry goods product selection site ,in Indicates the image data of the dry goods object selection at the dry goods product selection site The middle segment The image data of the selection result of a single dry goods object at the dry goods product selection site corresponding to each dry goods object; the image data of the selection result of a single dry goods object at the dry goods product selection site represents feedback image information for identifying the selection qualification analysis result of a single dry goods object at the dry goods product selection site.

[0040] Preferably, the steps of constructing the dry goods product selection on-site result image data and performing the dry goods product selection on-site result feedback operation are as follows:

[0041] S71: Collecting image data of the selection results of individual dry goods objects at the dry goods product selection site Image data of the selection results of a single dry goods object at the dry goods product selection site Image data of dry goods objects selected in sequence according to the number of dry goods objects and the dry goods product selection site In the on-site selection of dry goods products, the dry goods object frame selection image is combined with image feature matching to construct the dry goods product selection on-site result image data ;

[0042] S72: Generate image data of the on-site result of the dry goods product selection through the mobile terminal display screen Carry out feedback on on-site selection results of dry goods products.

[0043] An AI-based intelligent selection system for customers purchasing dry goods products, used to implement the AI-based intelligent selection method for customers purchasing dry goods products, the system includes a dry goods product selection search module, a dry goods product selection processing module, and a dry goods product selection feedback module;

[0044] The dry goods product selection and search module includes a dry goods product feature information collection unit, a dry goods product selection site image collection unit, a dry goods product single dry goods object frame selection and calibration image storage unit, a target dry goods product single dry goods object frame selection and calibration image search unit, a dry goods product qualified overall feature image library storage unit, and a target dry goods product qualified overall feature image library matching unit;

[0045] The dry goods product feature information collection unit collects target dry goods product feature text data through a mobile terminal; the dry goods product selection site image collection unit collects dry goods product selection site image data through a mobile terminal; the dry goods product single dry goods object frame selection calibration image storage unit is used to store dry goods product single dry goods object frame selection calibration image data; the target dry goods product single dry goods object frame selection calibration image search unit performs a single dry goods object frame selection calibration reference image search process in the target dry goods product selection site image based on the target dry goods product feature text data and the dry goods product single dry goods object frame selection calibration image data to generate the target dry goods product single dry goods object frame selection calibration image data; the dry goods product qualified overall feature image library storage unit is used to store a dry goods product qualified overall feature image library; the target dry goods product qualified overall feature image library matching unit performs a qualified dry goods product appearance feature image library required for target dry goods selection analysis based on the target dry goods product feature text data and the dry goods product qualified overall feature image library to generate the target dry goods product qualified overall feature image library;

[0046] The dry goods product selection processing module includes a dry goods object frame selection image generation unit at the dry goods product selection site, a single dry goods object frame selection image segmentation unit at the dry goods product selection site, a single dry goods object qualified result analysis unit at the dry goods product selection site, a dry goods object qualified analysis result identification watermark image storage unit, a dry goods product selection site single dry goods object qualified analysis result identification watermark image matching unit, and a dry goods product selection site single dry goods object selection result image generation unit;

[0047] The dry goods object frame selection image generation unit at the dry goods product selection site performs frame selection calibration processing on a single dry goods object in the target dry goods product selection site image based on the dry goods product selection site image data and the target dry goods product single dry goods object frame selection calibration image data, and generates dry goods object frame selection image data at the dry goods product selection site; the dry goods product selection site single dry goods object frame selection image segmentation unit performs single dry goods object image segmentation processing on the dry goods product selection site dry goods object frame selection image based on the dry goods product selection site dry object frame selection image data, and generates dry goods product selection site single dry goods object frame selection image data; the dry goods product selection site single dry goods object qualified result analysis unit performs single dry goods object selection quality qualified result analysis processing on the dry goods product selection site based on the dry goods product selection site single dry goods object frame selection image data and the target dry goods product qualified overall feature image library, and generates dry goods product selection site single dry goods object qualified result analysis data. the dry goods object qualified analysis result identification watermark image storage unit is used to store the dry goods object qualified analysis result identification watermark image data; the dry goods product selection site single dry goods object qualified analysis result identification watermark image matching unit performs identification watermark image matching processing required for the dry goods product selection site single dry goods object different selection quality qualified analysis results according to the dry goods product selection site single dry goods object qualified result analysis data and the dry goods object qualified analysis result identification watermark image data, and generates the dry goods product selection site single dry goods object qualified analysis result identification watermark image data; the dry goods product selection site single dry goods object selection result image generation unit performs selection result image identification processing of the dry goods product selection site single dry goods object according to the dry goods product selection site single dry goods object qualified analysis result identification watermark image data and the dry goods product selection site single dry goods object frame selection image data, and generates the dry goods product selection site single dry goods object selection result image data;

[0048] The dry goods product selection feedback module includes a dry goods product selection on-site selection result image construction unit and a dry goods product selection on-site selection result feedback unit;

[0049] The dry goods product selection on-site selection result image construction unit is used to construct dry goods product selection on-site result image data; the dry goods product selection on-site selection result feedback unit performs dry goods product selection on-site selection result feedback operations based on the dry goods product selection on-site result image data combined with the mobile terminal display screen.

[0050] (3) Beneficial effects

[0051] The present invention provides an AI-based intelligent selection system and method for customers to purchase dry goods products. It has the following beneficial effects:

[0052] 1. Conveniently and accurately obtain target dry goods product feature information and target dry goods product selection site image parameters through mobile terminals, providing reliable data support for the intelligent identification of qualified dry goods products; accurately search for dry goods object selection and calibration reference images in target dry goods product selection site images based on target dry goods product feature information combined with artificial intelligence recognition algorithms and scientifically preset dry goods product single dry goods object selection and calibration image parameters, thereby achieving accurate and efficient identification of dry goods object types at the dry goods product selection site and improving the reliability of dry goods product selection; autonomously match the qualified dry goods product appearance feature image library required for target dry goods product selection analysis based on target dry goods product feature information combined with intelligent search algorithms and standard-set dry goods product qualified overall feature image library, thereby achieving intelligent matching of qualified dry goods image libraries required for different types of dry goods products and improving the quality of dry goods product selection.

[0053] Second, through image analysis, all individual dry goods object images at the dry goods product selection site can be accurately identified and framed, and all individual dry goods object framed image information at the dry goods product selection site can be accurately segmented to achieve accurate extraction of feature image information of all individual dry goods objects at the dry goods product selection site; based on the framed image information of individual dry goods objects at the dry goods product selection site, combined with intelligent search algorithms and the qualified overall feature image library of target dry goods products, a comprehensive intelligent evaluation of the selection quality of individual dry goods objects at the dry goods product selection site can be performed, achieving efficient and accurate selection of dry goods products and improving the accuracy of dry goods product selection; based on the dry goods product selection The qualified result analysis parameters of the single dry goods object at the selection site and the watermark image information of the qualified analysis result identification of the dry goods object are scientifically matched with the quality identification watermark image of the single dry goods object at the dry goods product selection site, so as to realize the intuitive identification of the quality of the dry goods product selection; the selection result image of the single dry goods object at the dry goods product selection site is visualized based on the qualified analysis result identification watermark image parameters of the single dry goods object at the dry goods product selection site combined with image analysis and the frame selection image parameters of the single dry goods object at the dry goods product selection site, so as to realize the intuitive watermark image identification of the selection result of the single dry goods object at the dry goods product selection site, and improve the applicability of dry goods selection.

[0054] 3. By combining the image of the selection result of a single dry goods object at the dry goods product selection site with image feature analysis and the frame selection image of the dry goods object at the dry goods product selection site for image matching, the dry goods product selection site result image information is scientifically constructed, and the selection quality type of a single dry goods object at the dry goods product selection site is accurately marked, making it easier for customers to intuitively and accurately identify the quality of dry goods products; at the same time, combined with the mobile terminal display screen, the dry goods product on-site selection result feedback operation is independently and reliably executed, realizing convenient and efficient output of dry goods product selection results, and improving customers' dry goods purchasing experience and dry goods purchase quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1This is a module diagram of the AI-based intelligent selection system for customers purchasing dry goods products provided by the present invention;

[0056] Figure 2 This is a flowchart of the AI-based intelligent selection method for customers to purchase dry goods products provided by the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0058] The embodiments of the AI-based intelligent selection system and method for customers to purchase dry goods products are as follows:

[0059] Example 1:

[0060] See also Figure 1 - Figure 2 , an AI-based intelligent selection method for customers to purchase dry goods products, characterized in that the method includes the following steps:

[0061] S1. Collecting target dry goods product feature text data and dry goods product selection site image data;

[0062] S2. Performing a search process for a single dry goods object frame selection calibration reference image in the target dry goods product selection scene image based on the target dry goods product feature text data and the single dry goods object frame selection calibration image data of the dry goods product to generate the single dry goods object frame selection calibration image data of the target dry goods product;

[0063] S3. Perform matching processing on the appearance feature image library of qualified dry goods products required for target dry goods product selection and analysis based on the target dry goods product feature text data and the qualified overall feature image library of dry goods products to generate a qualified overall feature image library of target dry goods products;

[0064] S4. Performing a frame selection and calibration process on a single dry goods object in the target dry goods product selection scene image based on the dry goods product selection scene image data and the frame selection and calibration image data of a single dry goods object of the target dry goods product to generate frame selection image data of the dry goods product selection scene dry goods object, and performing a single dry goods object image segmentation process on the frame selection image of the dry goods product selection scene dry goods object to generate frame selection image data of a single dry goods object of the dry goods product selection scene dry goods object;

[0065] S5. Analyze and process the qualified results of the selection quality of the individual dry goods objects at the dry goods product selection site based on the frame selection image data of the individual dry goods objects at the dry goods product selection site and the qualified overall feature image library of the target dry goods products, thereby generating qualified result analysis data of the individual dry goods objects at the dry goods product selection site;

[0066] S6. Perform identification watermark image matching processing required for different selection quality qualified analysis results of individual dry goods objects at the dry goods product selection site based on the qualified result analysis data of individual dry goods objects at the dry goods product selection site and the identification watermark image data of the qualified analysis results of the dry goods objects, thereby generating the identification watermark image data of the qualified analysis results of individual dry goods objects at the dry goods product selection site, and perform selection result image identification processing of the individual dry goods objects at the dry goods product selection site on the frame selection image data of the individual dry goods objects at the dry goods product selection site, thereby generating the selection result image data of the individual dry goods objects at the dry goods product selection site;

[0067] S7. Construct image data of the on-site results of the dry goods product selection and perform a dry goods product on-site selection result feedback operation.

[0068] For further information, see Figure 1 - Figure 2 The steps for collecting target dry goods product feature text data and dry goods product selection scene image data are as follows:

[0069] S11. Input the characteristic text information of the dry goods product to be purchased by the customer online through the mobile terminal, and generate the characteristic text data of the target dry goods product The target dry goods product feature text data includes the name text data, variety text data, origin text data, and brand text data of the dry goods product; the mobile terminal includes any one of a smart phone, a smart watch, and a tablet computer;

[0070] Use mobile terminals to capture on-site image information of dry goods products that customers are waiting to purchase in an unpacked state, and generate on-site image data of dry goods product selection .

[0071] The steps for searching and processing a single dry product object frame selection calibration reference image in the target dry product selection scene image based on the target dry product feature text data and the dry product single dry product object frame selection calibration image data to generate the target dry product single dry product object frame selection calibration image data are as follows:

[0072] S21. Create a single dry product object selection calibration image data set , ;in Indicates the The image data of a single dry product object of a dry product corresponding to the dry product combination feature type is selected and calibrated. Indicates the maximum number of dry food product feature types; dry food product combination feature type represents an index identifier type formed by a combination of name, variety, origin, and brand feature information, used to search for a specific dry food product single dry food object selection calibration image; dry food product single dry food object selection calibration image data represents a reference image used for single dry food object image recognition and selection in scene images of different types of dry food products;

[0073] S22. Target dry goods product feature text data A collection of image data for selecting and calibrating a single dry product object Marking and calibrating image data for a single dry product object in a medium-sized dry product Perform dry goods product feature character matching to search for target dry goods product feature text data Corresponding dry goods product single dry goods object frame selection calibration image data , and generate the target dry goods product single dry goods object selection calibration image data through data identification , execute to generate target dry goods product single dry goods object selection calibration image data The specific steps are as follows:

[0074] S221, initialization, update the maximum number of iterations T and update the calibration image to search for the pelican population location. The formula for updating the calibration image to search for the pelican population location is as follows: ,in Indicates the The calibrated image searches for the pelican individual in The position of the dimension, that is, The calibration images search for pelican individuals in the dimension A collection of image data for a single dry product object selection and calibration The position in the search space, Indicates a random integer for position adjustment, rand indicates a random number in the range [0,1], and Indicates in The upper and lower boundaries of the problem are solved in the dimension A collection of image data for a single dry product object selection and calibration Search for text data related to the target product characteristics in the search space Matching dry goods products Single dry goods object frame selection calibration image data The number of upper and lower boundaries;

[0075] S222, exploration phase, calibration image search pelican to determine the location of prey, that is, in the dry goods product single dry goods object frame selection calibration image data set Search for text data related to the target product characteristics in the search space Matching dry goods products Single dry goods object frame selection calibration image data The location of the calibration image is then searched for the pelican individual to the dry goods product single dry goods object box to select the calibration image data The algorithm selects the calibration image data set for a single dry product object by moving the calibration image data set. Scan the search space and select and calibrate the image data of a single dry object of the target dry product The location of the prey in the dry product single dry object selection calibration image data set The search space is randomly generated, and the approximate prey strategy formula is as follows: ,in Indicates the exploration phase after the update The calibrated image searches for the pelican individual in The position of the dimension, that is, the position of the first dimension after the exploration phase update The calibration images search for pelican individuals in the dimension A collection of image data for a single dry product object selection and calibration Search for text data related to the target product characteristics in the search space Matching dry goods products Single dry goods object frame selection calibration image data location, For prey in The position of the dimension, that is, the target dry product feature text data Matching dry goods products Single dry goods object frame selection calibration image data The prey is in the dimension A collection of image data for a single dry product object selection and calibration The position in the search space, is the objective function value of the prey, For the The objective function value of the individual pelican is searched from the calibration images;

[0076] S223, Development Phase, Calibration Image Search Pelican selects calibration image data set for a single dry product object All dry goods products in the search space Single dry goods object selection calibration image data Perform target prey hunting search, that is, select and calibrate the image data set of a single dry object in the dry product Search for text data related to the target product characteristics in the search space Matching dry goods products Single dry goods object frame selection calibration image data , the behavior process of the calibration image search pelican is modeled. The algorithm checks the position near the calibration image search pelican position so that the algorithm converges to a better position. The calculation formula of the calibration image search pelican hunting behavior is as follows: ,in Indicates the update after the development phase The calibrated image search pelican The position of the dimension, that is, the first The calibrated images search Pelican in dimension A collection of image data for a single dry product object selection and calibration Search for text data related to the target product characteristics in the search space Matching dry goods products Single dry goods object frame selection calibration image data location, is a random integer of 0 or 2; t is the current number of iterations; T is the maximum number of iterations;

[0077] S224. When the algorithm meets the maximum number of iterations, output the target product feature text data Matching dry goods products Single dry goods object frame selection calibration image data ;

[0078] S225: Mark the image data of the dry goods product output in step S224 by selecting the single dry goods object Generate target dry goods product single dry goods object selection calibration image data after data identification .

[0079] The steps for matching the target dry goods product feature text data with the qualified overall feature image library of the dry goods products required for target dry goods product selection analysis to generate the qualified overall feature image library of the target dry goods products are as follows:

[0080] S31. Establish a collection of qualified overall feature image libraries for dry goods products ,in Indicates the A qualified overall feature image library of dry goods products corresponding to the combination feature types of dry goods products; , ;in Image library representing qualified overall features of dry goods products Middle A piece of qualified overall characteristic image data of dry goods products, Image library representing qualified overall features of dry goods products The maximum number of qualified overall feature images of dry food products; the qualified overall feature image data of dry food products represent the overall appearance image information of qualified dry food products established by different types of dry food product standards;

[0081] S32, use KD tree nearest neighbor search algorithm to target dry goods product feature text data A collection of image libraries with qualified overall features of dry goods products Image library of qualified overall features of medium-quality products Perform keyword matching on dry goods product features and search for target dry goods product feature text data Corresponding dry goods product qualified overall feature image library , and build a qualified overall feature image library of target dry goods products ,in Indicates the A piece of qualified overall characteristic image data of the target dry goods product.

[0082] Through the cooperation of the dry goods product feature information collection unit and the dry goods product selection site image collection unit, a mobile terminal is used to conveniently and accurately obtain the target dry goods product feature information and the target dry goods product selection site image parameters, providing reliable data support for the intelligent identification of qualified dry goods products; the target dry goods product single dry goods object frame selection calibration image search unit, based on the target dry goods product feature information combined with the artificial intelligence recognition algorithm and the scientifically preset dry goods product single dry goods object frame selection calibration image parameters, accurately searches for the dry goods object frame selection calibration reference image in the target dry goods selection site image, realizes accurate and efficient identification of the dry goods object type at the dry goods product selection site, and improves the reliability of dry goods product selection; the target dry goods product qualified overall feature image library matching unit, based on the target dry goods product feature information combined with the intelligent search algorithm and the standard set dry goods product qualified overall feature image library, performs autonomous matching of the qualified dry goods product appearance feature image library required for target dry product selection analysis, realizes intelligent matching of the qualified dry product image library required for different types of dry goods products, and improves the quality of dry goods product selection.

[0083] For further information, see Figure 1 - Figure 2 Based on the dry goods product selection scene image data and the target dry goods product single dry goods object frame selection calibration image data, a single dry goods object frame selection calibration process is performed in the target dry goods product selection scene image to generate dry goods product selection scene dry object frame selection image data and perform single dry goods object image segmentation processing on the dry goods product selection scene dry object frame selection image. The operation steps for generating the dry goods product selection scene single dry goods object frame selection image data are as follows:

[0084] S41. Use FLANN algorithm to select and calibrate image data based on the target dry goods product single dry goods object On-site image data for dry goods product selection The single dry goods object is identified in the corresponding dry goods product selection scene image, and the identified single dry goods object is calibrated using closed line frame selection in the dry goods product selection scene image, and the dry goods object frame selection image data of the dry goods product selection scene is generated. ;

[0085] S42, selecting image data of the dry goods object on the dry goods product selection site The framed image of a single dry goods object is segmented and extracted along the closed line of the frame calibration, and a single dry goods object framed image data set of the dry goods product selection site is generated. , ;in Indicates the image data of the dry goods object selection at the dry goods product selection site The middle segment The image data of a single dry goods object selected at the dry goods product selection site corresponding to each dry goods object, Indicates the image data of the dry goods object selection at the dry goods product selection site The maximum number of dry goods objects.

[0086] The steps for analyzing and processing the qualified results of the selection of a single dry goods object at the dry goods product selection site based on the frame selection image data of the single dry goods object at the dry goods product selection site and the qualified overall feature image library of the target dry goods product are as follows to generate the qualified result analysis data of the single dry goods object at the dry goods product selection site:

[0087] S51. Use FLANN algorithm to select image data sets of single dry goods objects at the dry goods product selection site Image data of a single dry goods object selected at the dry goods product selection site Ordered by the number of dry goods objects and the qualified overall feature image library of target dry goods products Qualified overall feature image data of target dry goods products Perform image feature matching and generate a data set of qualified results analysis of a single dry goods object at the dry goods product selection site based on the image feature matching results ,in Indicates the image data of the dry goods object selection at the dry goods product selection site The middle segment Analytical data on the qualified results of individual dry goods objects at the selection site for dry goods products corresponding to each dry goods object;

[0088] when and The image feature matching is successful, indicating that The quality of each dry goods product is qualified and meets the selection criteria; then the qualified result analysis data of each dry goods object at the dry goods product selection site is output. To be qualified;

[0089] when and No image features were matched successfully, indicating that If the quality of a dry goods product fails to meet the selection criteria, the qualified result analysis data of a single dry goods object at the dry goods product selection site will be output. Is unqualified.

[0090] Based on the qualified result analysis data of a single dry goods object at the dry goods product selection site and the dry goods object qualified analysis result identification watermark image data, identification watermark image matching processing required for the qualified analysis results of different selection qualities of the single dry goods object at the dry goods product selection site is performed to generate the qualified analysis result identification watermark image data of the single dry goods object at the dry goods product selection site, and the selection result image identification processing of the single dry goods object at the dry goods product selection site is performed on the single dry goods object frame selection image data at the dry goods product selection site to generate the selection result image data of the single dry goods object at the dry goods product selection site as follows:

[0091] S61. Establishing a watermark image data set for qualified analysis results of dry goods objects ,in Indicates the qualified result analysis data of a single dry goods object at the dry goods product selection site The watermark image data of the qualified analysis result of the corresponding dry goods object is qualified. At this time, the watermark image of the qualified analysis result of the dry goods object is a watermark image with the word "qualified"; Indicates the qualified result analysis data of a single dry goods object at the dry goods product selection site The watermark image data of the qualified analysis result of the corresponding dry goods object when it is unqualified, in this case, the watermark image of the qualified analysis result of the dry goods object is a watermark image with the word "unqualified";

[0092] S62, using a bidirectional search algorithm to analyze the qualified results of a single dry goods object on the dry goods product selection site Analysis data of qualified results of individual dry goods objects at the selection site of medium dry goods products A watermarked image data set is numbered and ordered according to the number of dry goods objects and marked with the qualified analysis results of the dry goods objects. The qualified analysis result watermark image data of the dry goods object is used to match the keyword of the qualified analysis result of the single dry goods object at the dry goods product selection site, and the qualified analysis data of the single dry goods object at the dry goods product selection site is searched out. The corresponding dry goods object qualified analysis result identification watermark image data, and generate a dry goods product selection site single dry goods object qualified analysis result identification watermark image data set ,in Indicates the image data of the dry goods object selection at the dry goods product selection site The middle segment Watermark image data of qualified analysis results of a single dry goods object at the dry goods product selection site corresponding to each dry goods object;

[0093] S63, marking the watermark image data set of the qualified analysis result of the single dry goods object at the dry goods product selection site Watermark image data of qualified analysis results of a single dry goods object at the dry goods product selection site Image data set of single dry goods object selection based on dry goods object quantity number and dry goods product selection site Image data of a single dry goods object selected at the dry goods product selection site Combine images and generate a dataset of image data of individual dry goods object selection results at the dry goods product selection site ,in Indicates the image data of the dry goods object selection at the dry goods product selection site The middle segment The image data of the selection result of a single dry goods object at the dry goods product selection site corresponding to each dry goods object; the image data of the selection result of a single dry goods object at the dry goods product selection site represents feedback image information used to identify the selection qualification analysis result of a single dry goods object at the dry goods product selection site.

[0094] Through the cooperation of the dry goods object frame selection image generation unit at the dry goods product selection site and the single dry goods object frame selection image segmentation unit at the dry goods product selection site, all single dry goods object images at the dry goods product selection site are accurately identified and framed based on image analysis, and all single dry goods object frame selection image information at the dry goods product selection site is accurately segmented, thereby accurately extracting the feature image information of all single dry goods objects at the dry goods product selection site; the single dry goods object qualified result analysis unit at the dry goods product selection site performs a comprehensive intelligent evaluation of the selection quality of single dry goods objects at the dry goods product selection site based on the single dry goods object frame selection image information at the dry goods product selection site combined with an intelligent search algorithm and a qualified overall feature image library of target dry goods products, thereby achieving efficient and accurate selection of dry goods products and improving the accuracy of dry goods product selection; The watermark image matching unit for identifying the qualified analysis result of a single dry goods object at the dry goods product selection site performs scientific matching of the watermark image for identifying the quality of the single dry goods object at the dry goods product selection site based on the qualified analysis result analysis parameters of the single dry goods object at the dry goods product selection site and the watermark image information of the qualified analysis result identification of the dry goods object, thereby realizing intuitive identification of the quality of the dry goods product selection; the selection result image generation unit for a single dry goods object at the dry goods product selection site performs visual identification of the selection result image of a single dry goods object at the dry goods product selection site based on the qualified analysis result identification watermark image parameters of the single dry goods object at the dry goods product selection site combined with image analysis and the frame selection image parameters of the single dry goods object at the dry goods product selection site, thereby realizing intuitive watermark image identification of the selection result of a single dry goods object at the dry goods product selection site, thereby improving the applicability of dry goods selection.

[0095] For further information, see Figure 1 - Figure 2 The steps for constructing the image data of the on-site selection results of dry goods products and performing the on-site selection result feedback task of dry goods products are as follows:

[0096] S71, collecting image data of the selection results of individual dry goods objects at the dry goods product selection site Image data of the selection results of a single dry goods object at the dry goods product selection site Order by number of dry goods objects and dry goods product selection on-site dry goods object selection image data In the on-site selection of dry goods products, the dry goods object frame selection image is combined with image feature matching to construct the dry goods product selection on-site result image data ;

[0097] S72, generating dry goods product selection on-site result image data through the mobile terminal display screen Carry out feedback on on-site selection results of dry goods products.

[0098] Through the cooperation of the dry goods product selection site selection result image construction unit and the dry goods product selection site selection result feedback unit, the selection result image of a single dry goods object at the dry goods product selection site is combined with image feature analysis and the dry goods object frame selection image at the dry goods product selection site for image matching and combination to scientifically construct the dry goods product selection site selection result image information, so as to achieve accurate marking of the selection quality type of a single dry goods object at the dry goods product selection site, so that customers can intuitively and accurately identify the quality of dry goods products; at the same time, combined with the mobile terminal display screen, the dry goods product selection site selection result feedback operation is independently and reliably executed, so as to achieve convenient and efficient output of dry goods product selection results, and improve customers' dry goods purchasing experience and dry goods purchase quality.

[0099] Example 2:

[0100] See also Figure 1 - Figure 2 , an AI-based intelligent selection system for customers to purchase dry goods products, which is used to implement an AI-based intelligent selection method for customers to purchase dry goods products. The system includes a dry goods product selection search module, a dry goods product selection processing module, and a dry goods product selection feedback module;

[0101] The dry goods product selection and search module includes a dry goods product feature information collection unit, a dry goods product selection site image collection unit, a dry goods product single dry goods object frame selection and calibration image storage unit, a target dry goods product single dry goods object frame selection and calibration image search unit, a dry goods product qualified overall feature image library storage unit, and a target dry goods product qualified overall feature image library matching unit;

[0102] a dry goods product feature information collection unit, which collects target dry goods product feature text data through a mobile terminal; a dry goods product selection site image collection unit, which collects dry goods product selection site image data through a mobile terminal; a dry goods product single dry goods object frame selection calibration image storage unit, which is used to store dry goods product single dry goods object frame selection calibration image data; a target dry goods product single dry goods object frame selection calibration image search unit, which performs frame selection calibration reference image search processing for a single dry goods object in the target dry goods selection site image based on the target dry goods product feature text data and the dry goods product single dry goods object frame selection calibration image data, and generates target dry goods product single dry goods object frame selection calibration image data; a dry goods product qualified overall feature image library storage unit, which is used to store a dry goods product qualified overall feature image library; a target dry goods product qualified overall feature image library matching unit, which performs matching processing on a qualified dry goods product appearance feature image library required for target dry product selection analysis based on the target dry goods product feature text data and the dry goods product qualified overall feature image library, and generates a target dry goods product qualified overall feature image library;

[0103] The dry goods product selection processing module includes a dry goods object frame selection image generation unit at the dry goods product selection site, a single dry goods object frame selection image segmentation unit at the dry goods product selection site, a single dry goods object qualified result analysis unit at the dry goods product selection site, a dry goods object qualified analysis result identification watermark image storage unit, a dry goods product selection site single dry goods object qualified analysis result identification watermark image matching unit, and a dry goods product selection site single dry goods object selection result image generation unit;

[0104] A dry goods object frame selection image generation unit at the dry goods product selection site performs frame selection calibration processing on a single dry goods object in the target dry goods product selection site image based on the dry goods product selection site image data and the target dry goods product single dry goods object frame selection calibration image data, and generates dry goods product selection site dry goods object frame selection image data; a dry goods product selection site single dry goods object frame selection image segmentation unit performs single dry goods object image segmentation processing on the dry goods product selection site dry goods object frame selection image based on the dry goods product selection site dry object frame selection image data, and generates dry goods product selection site single dry goods object qualified result analysis unit, performs single dry goods object selection quality qualified result analysis processing on the dry goods product selection site single dry goods object based on the dry goods product selection site single dry object frame selection image data and the target dry goods product qualified overall feature image library, and generates dry goods product selection site single dry goods object qualified result analysis data; a dry goods object qualified analysis result identification watermark image storage unit, used to store dry goods object qualified analysis result identification watermark image data; a dry goods product selection site single dry goods object qualified analysis result identification watermark image matching unit, which performs identification watermark image matching processing required for qualified analysis results of different selection qualities of single dry goods objects at the dry goods product selection site based on the dry goods product selection site single dry goods object qualified result analysis data and the dry goods object qualified analysis result identification watermark image data, and generates dry goods product selection site single dry goods object qualified analysis result identification watermark image data; a dry goods product selection site single dry goods object selection result image generation unit, which performs selection result image identification processing of single dry goods objects at the dry goods product selection site based on the dry goods product selection site single dry goods object qualified analysis result identification watermark image data and the dry goods product selection site single dry goods object frame selection image data, and generates dry goods product selection site single dry goods object selection result image data;

[0105] The dry goods product selection feedback module includes a dry goods product selection on-site selection result image construction unit and a dry goods product selection on-site selection result feedback unit;

[0106] The dry goods product selection on-site selection result image construction unit is used to construct the dry goods product selection on-site result image data; the dry goods product selection on-site selection result feedback unit performs the dry goods product selection on-site selection result feedback operation based on the dry goods product selection on-site result image data combined with the mobile terminal display screen.

[0107] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An AI-based intelligent selection method for customers to purchase dry goods products, characterized by: The method comprises the following steps: S1. Collect target product feature text data and dry goods product selection on-site image data ; S2. Performing a search process for a single dry goods object frame selection calibration reference image in the target dry goods product selection scene image based on the target dry goods product feature text data and the dry goods product single dry goods object frame selection calibration image data to generate the target dry goods product single dry goods object frame selection calibration image data; The S2 comprises the following steps: S21. Create a single dry product object selection calibration image data set , ;in Indicates the The image data of a single dry product object of a dry product corresponding to the dry product combination feature type is selected and calibrated. Indicates the maximum number of dry goods product feature types; S22, the With the As stated in Perform dry goods product feature character matching and search for the The corresponding , and generate the target dry goods product single dry goods object selection calibration image data through data identification : S3. Performing matching processing on the appearance feature image library of qualified dry goods products required for target dry goods product selection and analysis based on the target dry goods product feature text data and the qualified overall feature image library of dry goods products to generate a qualified overall feature image library of target dry goods products; The S3 includes the following steps: S31. Establish a collection of qualified overall feature image libraries for dry goods products ,in Indicates the A qualified overall feature image library of dry goods products corresponding to the combination feature types of dry goods products; , ;in Indicates the Middle A piece of qualified overall characteristic image data of dry goods products, Indicates the The maximum number of qualified overall feature images for medium-quality products; S32, using KD tree nearest neighbor search algorithm to With the As stated in Perform keyword matching of dry goods product features and search for the The corresponding , and build a qualified overall feature image library of target dry goods products ,in Indicates the Qualified overall characteristic image data of target dry goods products; S4. Performing a framing and calibration process on a single dry goods object in the target dry goods product selection scene image based on the dry goods product selection scene image data and the framing and calibration image data of a single dry goods object of the target dry goods product to generate framing and calibration image data of the dry goods product selection scene dry goods object, and performing a framing and segmentation process on the single dry goods object image of the dry goods product selection scene dry goods object to generate framing and calibration image data of the single dry goods object of the dry goods product selection scene dry goods object; The S4 comprises the following steps: S41, using FLANN algorithm according to In the The single dry goods object is identified in the corresponding dry goods product selection scene image, and the identified single dry goods object is calibrated using closed line frame selection in the dry goods product selection scene image, and the dry goods object frame selection image data of the dry goods product selection scene is generated. ; S42, the The framed image of a single dry goods object is segmented and extracted along the closed line of the frame calibration, and a single dry goods object framed image data set of the dry goods product selection site is generated. , ;in Indicates the The middle segment The image data of a single dry goods object selected at the dry goods product selection site corresponding to each dry goods object, Indicates the The maximum number of dry goods objects; S5. Analyze and process the qualified results of the selection of the individual dry goods object at the dry goods product selection site based on the frame selection image data of the individual dry goods object at the dry goods product selection site and the qualified overall feature image library of the target dry goods product, to generate qualified result analysis data of the individual dry goods object at the dry goods product selection site; The S5 comprises the following steps: S51, using FLANN algorithm to As stated in According to the number of dry goods objects, they are numbered and arranged in order. As stated in Perform image feature matching and generate a data set of qualified results analysis of a single dry goods object at the dry goods product selection site based on the image feature matching results ,in Indicates the The middle segment Analytical data on the qualified results of individual dry goods objects at the selection site for dry goods products corresponding to each dry goods object; when and If the image feature matching is successful, the qualified result analysis data of a single dry goods object at the dry goods product selection site will be output. To be qualified; when and If no image features are matched successfully, the qualified result analysis data of a single dry goods object at the dry goods product selection site will be output. is unqualified; S6. Perform identification watermark image matching processing required for different selection quality qualified analysis results of individual dry goods objects at the dry goods product selection site based on the qualified result analysis data of the individual dry goods objects at the dry goods product selection site and the identification watermark image data of the qualified analysis results of the dry goods objects, generate the identification watermark image data of the qualified analysis results of the individual dry goods objects at the dry goods product selection site, and perform selection result image identification processing of the individual dry goods objects at the dry goods product selection site on the frame selection image data of the individual dry goods objects at the dry goods product selection site, generate the selection result image data of the individual dry goods objects at the dry goods product selection site; The S6 comprises the following steps: S61. Establishing a watermark image data set for qualified analysis results of dry goods objects ,in Indicates the The watermark image data of the qualified analysis result of the corresponding dry goods object is qualified. At this time, the watermark image of the qualified analysis result of the dry goods object is a watermark image with the word "qualified"; Indicates the The watermark image data of the qualified analysis result of the corresponding dry goods object when it is unqualified, in this case, the watermark image of the qualified analysis result of the dry goods object is a watermark image with the word "unqualified"; S62, using a bidirectional search algorithm to As stated in According to the number of dry goods objects, they are numbered and arranged in order. The qualified analysis result of the dry goods object is identified by the watermark image data for the dry goods product selection site, and the qualified analysis result of the single dry goods object is matched with the keyword to search for the The corresponding dry goods object qualified analysis result identification watermark image data, and generate a dry goods product selection site single dry goods object qualified analysis result identification watermark image data set ,in Indicates the The middle segment Watermark image data of qualified analysis results of a single dry goods object at the dry goods product selection site corresponding to each dry goods object; S63, the As stated in According to the number of dry goods objects and the As stated in Combine images and generate a dataset of image data of individual dry goods object selection results at the dry goods product selection site ,in Indicates the The middle segment Image data of a single dry goods object selection result at a dry goods product selection site corresponding to each dry goods object; S7. Construct image data of the on-site results of the dry goods product selection and perform a dry goods product on-site selection result feedback operation.

2. The AI-based intelligent selection method for dry goods purchased by customers according to claim 1 is characterized by: Said S1 comprises the following steps: S11. Input the characteristic text information of the dry goods product to be purchased by the customer online through the mobile terminal, and generate the characteristic text data of the target dry goods product ; Use mobile terminals to capture on-site image information of dry goods products that customers are waiting to purchase in an unpacked state, and generate on-site image data of dry goods product selection .

3. The AI-based intelligent selection method for dry goods purchased by customers according to claim 1 is characterized by: The S22 includes the following steps: S221, initialization, updating the maximum number of iterations T and updating the calibration image to search for the location of the pelican population; S222, exploration phase, calibration image search pelican to determine the location of the prey, that is, in the Search the search space for Matching the The image is then calibrated to search for individual pelicans located in the The algorithm moves in the area and models the approaching prey strategy of the pelican in the calibration image search. The search space is scanned, and the target The location of the prey is described in The search space is randomly generated; S223, development stage, calibration image search pelican All the above in the search space Conducting target hunting search, i.e. Search the search space for Matching the ,This behavior process of searching the pelican in the calibration image is modeled, and the algorithm checks the positions near the ,position of the pelican in the calibration image search, so that the algorithm converges to a ,better position; S224, when the algorithm meets the maximum number of iterations, the output is the same as the Matching the ; S225, the output in step S224 Generate target dry goods product single dry goods object selection calibration image data after data identification .

4. The AI-based intelligent selection method for customers to purchase dry goods products according to claim 1 is characterized by: The S7 comprises the following steps: S71, the As stated in According to the number of dry goods objects, they are numbered and arranged in order. In the on-site selection of dry goods products, the dry goods object frame selection image is combined with image feature matching to construct the dry goods product selection on-site result image data ; S72, generating the Carry out feedback on on-site selection results of dry goods products.

5. An AI-based intelligent selection system for customers purchasing dry goods products, used to implement the AI-based intelligent selection method for customers purchasing dry goods products according to any one of claims 1 to 4, characterized in that: The system includes a dry goods product selection search module, a dry goods product selection processing module, and a dry goods product selection feedback module.

Images