Icon matching method and device, computer device and storage medium

Abstract

The embodiment of the application belongs to the artificial intelligence technology in the financial field, and relates to an icon matching method, which comprises the following steps: obtaining an icon recognition model trained and a business icon set, performing icon recognition on the icons in the business icon set by using the icon recognition model, and obtaining a business icon label set; storing the business icon label set in an icon library; obtaining a to-be-matched business icon, and judging whether the to-be-matched business icon matches the icon library; when the to-be-matched business icon matches the icon library, selecting the matched icon label in the icon library as a target icon label of the to-be-matched business icon; and when the to-be-matched business icon does not match the icon library, adding the icon label of the to-be-matched business icon in the business. The application also provides an icon matching device, a computer device and a storage medium. In addition, the application also relates to the blockchain technology, and the business icon set can be stored in the blockchain. The application improves the accuracy of icon matching and the efficiency of business updating and upgrading.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more particularly to an icon matching method, apparatus, computer device, and storage medium. Background Technology

[0002] With the development of internet technology, artificial intelligence image recognition tools can greatly improve the efficiency of human production and life. For example, facial recognition, vehicle recognition, and animal recognition are widely used in various internet application scenarios. In the financial sector, banks' various business systems also involve artificial intelligence image recognition technology during business updates and upgrades. However, during these updates and upgrades, the front-end business often involves thousands of front-end pages containing numerous icons, resulting in relatively low efficiency. Therefore, improving the efficiency of business update and upgrades has become an urgent problem to be solved.

[0003] There is a method for business update and upgrade that compiles and transmits a large number of duplicate files during the business update and upgrade process, so that each icon or image can be uniquely identified, thereby effectively managing all icons or images with unique identifiers.

[0004] However, the applicant discovered that traditional business update and upgrade methods have certain drawbacks. On the one hand, some front-end businesses often include tens of thousands of front-end pages during the update and upgrade process. These pages contain a large number of icons, and the same icons and images are frequently used during the business update and upgrade process. If the same icons are added to the business in every update and upgrade, it is easy to cause a large number of duplicate files to accumulate in the business. Furthermore, the business update and upgrade process requires the compilation and transmission of a large number of duplicate files, which reduces the efficiency of the business update and upgrade. On the other hand, since the update and upgrade of large-scale businesses often involves many developers and modules, a lot of manpower is required for icon matching during the business update and upgrade process, which leads to a decrease in the accuracy of icon matching during the business update and upgrade process. It can be seen that traditional business update and upgrade methods have the problems of low icon matching accuracy and low business update and upgrade efficiency. Summary of the Invention

[0005] The purpose of this application is to provide an icon matching method, apparatus, computer device, and storage medium, the main purpose of which is to improve the accuracy of icon matching and the efficiency of business updates and upgrades.

[0006] To address the aforementioned technical problems, this application provides an icon matching method, employing the following technical solution:

[0007] Obtain the trained icon recognition model and business icon set, and use the icon recognition model to perform icon recognition on the icons in the business icon set to obtain the business icon tag set;

[0008] Store the set of business icon tags in a preset icon library;

[0009] Obtain the business icon to be matched, and determine whether the business icon to be matched matches the icon library;

[0010] When the business icon to be matched matches the icon library, the matching icon mark in the icon library is selected as the target icon mark of the business icon to be matched.

[0011] When the icon to be matched does not match the icon library, an icon tag for the icon to be matched is added to the business.

[0012] Furthermore, determining whether the business icon to be matched matches the icon library includes:

[0013] Obtain the initial icon marker of the business icon to be matched, as well as the local icon marker and remote icon marker in the icon library;

[0014] Determine whether the initial icon marker matches the local icon marker;

[0015] When the initial icon tag matches the local icon tag, it is determined that the business icon to be matched matches the icon library;

[0016] If the initial icon tag does not match the local icon tag, then determine whether the initial icon tag matches the remote icon tag;

[0017] When the initial icon tag matches the remote icon tag, it is determined that the business icon to be matched matches the icon library;

[0018] If the initial icon tag does not match the remote icon tag, then it is determined that the business icon to be matched does not match the icon library.

[0019] Furthermore, selecting a matching icon from the icon library as the target icon for the business icon to be matched includes:

[0020] Obtain the business requirements for the business icon to be matched, and the list of icons in the icon library that match the business icon to be matched;

[0021] Select the target icon marker of the business icon to be matched from the icon list according to the business requirements.

[0022] Furthermore, the step of adding the icon marker for the business icon to be matched in the business includes:

[0023] Identify the icon category of the business icon to be matched;

[0024] The business path of the business icon to be matched is determined based on the icon category;

[0025] The icon of the business to be matched is added to the business based on the business path.

[0026] Furthermore, storing the business icon set into a preset icon library includes:

[0027] Obtain the icon path of the business icon tag set and identify the business category to which the icons in the business icon tag set belong;

[0028] The business icon set is stored in the corresponding icon library according to the icon path and the business category.

[0029] Furthermore, the step of using the icon recognition model to perform icon recognition on the icons in the business icon set to obtain a business icon tag set includes:

[0030] The icon feature set is obtained by performing convolution operations on the icons in the business icon set using the convolutional layer of the icon recognition model.

[0031] Max pooling is performed on the icon feature set using the pooling layer of the icon recognition model to obtain a pooled icon set;

[0032] The pooled icon set is input into the activation function of the fully connected layer in the icon recognition model, and the business icon tag set is output.

[0033] Furthermore, before obtaining the trained icon recognition model and business icon set, the method further includes:

[0034] Obtain the training icon set and the preset model training framework;

[0035] Configure the icon modeling workflow of the training icon set in the visualization modeling module of the model training framework, and create an initial icon recognition model according to the icon modeling workflow;

[0036] Obtain the real icon label set corresponding to the training icon set, input the training icon set into the initial icon recognition model, and obtain the predicted icon label set of the training icon set;

[0037] Calculate the loss value between the predicted icon label set and the real icon label set, adjust the parameters of the initial icon recognition model according to the loss value, until the loss value meets a preset threshold, and obtain the trained icon recognition model.

[0038] To address the aforementioned technical problems, this application also provides an icon matching device, which employs the following technical solution:

[0039] The recognition module is used to acquire the trained icon recognition model and business icon set, and to use the icon recognition model to recognize the icons in the business icon set to obtain a business icon tag set.

[0040] A storage module is used to store the business icon mark set into a preset icon library;

[0041] The matching module is used to obtain the business icon to be matched and determine whether the business icon to be matched matches the icon library;

[0042] The selection module is used to select the matching icon mark in the icon library as the target icon mark of the business icon to be matched when the business icon to be matched matches the icon library.

[0043] A new module is added to the business if the icon to be matched does not match the icon library, and then adds an icon tag for the icon to be matched to the business.

[0044] To address the aforementioned technical problems, this application also provides a computer device that employs the following technical solution:

[0045] Memory, storing at least one computer program; and

[0046] The processor executes the computer program stored in the memory to implement the icon matching method described above.

[0047] To address the aforementioned technical problems, this application also provides a computer-readable storage medium, employing the technical solution described below:

[0048] The computer-readable storage medium stores at least one computer program, which is executed by a processor in an electronic device to implement the icon matching method described above.

[0049] Compared with the prior art, the embodiments of this application have the following main advantages:

[0050] In this embodiment, after obtaining the trained icon recognition model and business icon set, the icon recognition model is first used to recognize the icons in the business icon set to obtain a business icon tag set. This allows for icon recognition of all business icons during business updates and upgrades, eliminating the need for secondary recognition of already recognized icons and improving icon recognition efficiency. Secondly, by storing the business icon tag set in a preset icon library, the same icons can be reused during business updates and upgrades, improving efficiency. Finally, by determining whether the business icon to be matched matches the icon library, the problem of requiring significant manpower for icon matching during business updates and upgrades can be solved, thereby improving the accuracy of icon matching during business updates. When a match is found, the matching icon tag in the icon library is selected as the target icon tag for the business icon to be matched. This not only achieves icon deduplication and improves the efficiency of business updates and upgrades but also enables automatic icon matching, further improving the accuracy of icon matching. When a mismatch is found, an icon tag for the business icon to be matched is added to the business. This improves the icon library during icon matching, further enhancing the accuracy of icon matching. Therefore, the icon matching method, apparatus, computer equipment, and storage medium proposed in this application can improve the accuracy of icon matching and the efficiency of business updates and upgrades. Attached Figure Description

[0051] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is an exemplary system architecture diagram to which this application can be applied;

[0053] Figure 2 A flowchart of an embodiment of the icon matching method according to this application;

[0054] Figure 3 yes Figure 2 A flowchart of a specific implementation of step S203;

[0055] Figure 4 This is a schematic diagram of the structure of one embodiment of the icon matching device according to this application;

[0056] Figure 5 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0058] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0059] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0060] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0061] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.

[0062] Terminal devices 101, 102, and 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, e-book readers, MP3 players (Moving Picture Experts Audio Layer III), MP4 players (Moving Picture Experts Audio Layer IV), laptops, and desktop computers, etc.

[0063] Server 105 can be a server that provides various services, such as a backend server that supports the pages displayed on terminal devices 101, 102, and 103.

[0064] It should be noted that the icon matching method provided in this application embodiment is generally executed by a server / terminal device, and correspondingly, the icon matching device is generally set in the server / terminal device.

[0065] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0066] Continue to refer to Figure 2 The diagram illustrates a flowchart of an embodiment of the icon matching method according to this application. The icon matching method includes the following steps:

[0067] S201. Obtain the trained icon recognition model and business icon set, and use the icon recognition model to perform icon recognition on the icons in the business icon set to obtain a business icon tag set.

[0068] In this embodiment of the invention, the business icon set refers to a set of icons collected based on actual business scenarios, including currency icons, stock price chart icons, and insurance icons (such as property insurance icons and life insurance icons). For example, in the financial field, if there is a property insurance client update and upgrade service, then the business icon set includes car insurance product icons, accident and health insurance product icons, and car insurance product closing icons, etc.

[0069] In this embodiment of the invention, the business icon mark set refers to the mark set corresponding to each business icon. For example, in insurance sales, the business icon mark set may include car insurance icon marks, engineering insurance icon marks, accident and health insurance icon marks, a close button, a more button, left arrows, right arrows, a phone book icon, a speaker icon, etc.

[0070] In this embodiment of the invention, the trained icon recognition model can be a CNN neural network model, including convolutional layers, pooling layers, and fully connected layers.

[0071] In this embodiment of the invention, by acquiring a trained icon recognition model and a set of business icons, the icon recognition model is used to recognize the icons in the set of business icons to obtain a set of business icon tags. This enables icon recognition of all business icons during business updates and upgrades without the need for secondary recognition of already recognized icons, thus improving icon recognition efficiency.

[0072] As an embodiment of the present invention, before obtaining the trained icon recognition model and business icon set, the method further includes:

[0073] Obtain the training icon set and the preset model training framework;

[0074] Configure the icon modeling workflow of the training icon set in the visualization modeling module of the model training framework, and create an initial icon recognition model according to the icon modeling workflow;

[0075] Obtain the real icon label set corresponding to the training icon set, input the training icon set into the initial icon recognition model, and obtain the predicted icon label set of the training icon set;

[0076] Calculate the loss value between the predicted icon label set and the real icon label set, adjust the parameters of the initial icon recognition model according to the loss value, until the loss value meets a preset threshold, and obtain the trained icon recognition model.

[0077] The model training framework can be TensorFlow, which can be used to train icon recognition models that meet different business needs.

[0078] In one embodiment of the present invention, the visualization modeling module can be the visualization canvas of PAI-Designer in TensorFlow. The canvas includes multiple node components, and users can customize and select different node components according to actual business needs to configure the icon model workflow. The workflow is configured by dragging and dropping different node components onto the visualization canvas and connecting the selected node components from top to bottom. Furthermore, running the configured icon model workflow on the visualization canvas yields an initial icon recognition model.

[0079] For example, if a property insurance client needs to be updated and upgraded, and this update and upgrade requires the addition of icon recognition, then the following components can be selected in sequence: property insurance product icon data reading component, convolution component in deep learning, pooling component in deep learning, fully connected component in deep learning, and icon binary classification component. By connecting and running the selected node components in sequence, an initial icon recognition model can be obtained.

[0080] Furthermore, in this embodiment of the invention, the training icon set is input into the initial icon recognition model, and the initial icon recognition model performs convolutional pooling operations on the training icons to obtain icon features; then, the icon features are spatially flattened through the fully connected layer of the model and input into the activation function to obtain predicted icon labels; further, the loss value between the predicted icon label set and the real icon label set is calculated through the Loss function; when the loss value is not less than a preset threshold (e.g., 0.8), the parameters of the initial icon recognition model are adjusted; when the loss value is less than the preset threshold (e.g., 0.8), the trained icon recognition model is obtained.

[0081] As an embodiment of the present invention, the step of using the icon recognition model to perform icon recognition on the icons in the business icon set to obtain a business icon tag set includes:

[0082] The icon feature set is obtained by performing convolution operations on the icons in the business icon set using the convolutional layer of the icon recognition model.

[0083] Max pooling is performed on the icon feature set using the pooling layer of the icon recognition model to obtain a pooled icon set;

[0084] The pooled icon set is input into the activation function of the fully connected layer in the icon recognition model, and the business icon tag set is output.

[0085] The convolution operation extracts the icon feature set from the business icon set using a 3x3 convolution kernel. By performing a 2x2 max pooling operation on the icon feature set, the most important features in the icon feature set can be retained while some useless features are removed. This improves the computation speed while ensuring the accuracy of the extracted icon features, thereby achieving both efficiency and accuracy in icon recognition.

[0086] In one embodiment of the present invention, the pooled icon set is input into a fully connected layer and reshaped into a row. Finally, the ReLU activation function is used to perform a Dropout operation on the reshaped data to obtain the business icon tag set.

[0087] S202. Store the business icon set in a preset icon library.

[0088] In this embodiment of the invention, the preset icon library refers to a database in which the enterprise stores icon data. The icon library can be a traditional data warehouse or a user-defined file storage repository. When the data volume is large, a data warehouse is usually used for storage to improve storage speed. The icon library includes a remote icon library and a local icon library. The remote icon library includes a public online icon CDN library commonly used within the enterprise, and icons for all business operations of the enterprise can be stored in the remote icon library. The local icon library only includes icons for the current business.

[0089] In this embodiment of the invention, by storing the business icon mark set in a preset icon library, the same icons can be reused during business updates and upgrades. This solves the problem that adding the same icons to the business in every business update and upgrade easily leads to a large number of duplicate files accumulating in the business, and that a large number of duplicate files need to be compiled and transmitted during the business update and upgrade process, thus improving the efficiency of business updates and upgrades.

[0090] As an embodiment of the present invention, storing the business icon mark set into a preset icon library includes:

[0091] Obtain the icon path of the business icon tag set and identify the business category to which the icons in the business icon tag set belong;

[0092] The business icon set is stored in the corresponding icon library according to the icon path and the business category.

[0093] The icon path refers to the path of the current icon in the business icon mark set; the business category includes local business (i.e., the current business) or remote business. When the business category is identified as local business, the business icon mark is stored in the local icon library in the icon library; when the business category is identified as remote business, the business icon mark is stored in the remote icon library in the icon library.

[0094] In one embodiment of the present invention, by storing the business icon mark set in a local icon library or a remote icon library, it can be ensured that the business icon mark can be used in all business operations of the enterprise without repeated storage, thereby improving the efficiency of business updates and upgrades.

[0095] S203. Obtain the business icon to be matched, and determine whether the business icon to be matched matches the icon library.

[0096] In this embodiment of the invention, the business icon to be matched refers to a business icon that needs to be added during the business update and upgrade process. The business icon to be matched can be obtained by receiving an entry command from a user through a sidebar floating window on the business front-end page, and then retrieving the business icon uploaded by the pop-up layer on the business front-end page based on the entry command.

[0097] In this embodiment of the invention, by determining whether the business icon to be matched matches the icon library, the problem of needing to spend a lot of manpower on icon matching during business updates and upgrades can be solved, thereby improving the accuracy of icon matching.

[0098] As an embodiment of the present invention, reference Figure 3 As shown, determining whether the business icon to be matched matches the icon library includes the following steps S2031-S2036:

[0099] S2031. Obtain the initial icon marker of the business icon to be matched, as well as the local icon marker and remote icon marker in the icon library;

[0100] S2032. Determine whether the initial icon mark matches the local icon mark;

[0101] S2033. When the initial icon tag matches the local icon tag, it is determined that the business icon to be matched matches the icon library;

[0102] S2034. If the initial icon mark does not match the local icon mark, then determine whether the initial icon mark matches the remote icon mark;

[0103] S2035. When the initial icon tag matches the remote icon tag, it is determined that the business icon to be matched matches the icon library;

[0104] S2036. If the initial icon marker does not match the remote icon marker, then it is determined that the business icon to be matched does not match the icon library.

[0105] Specifically, by identifying whether the local icon markers are stored in a traditional data warehouse or a user-defined file repository, if the local icon markers are stored in a traditional data warehouse, an SQL query can be used to search for the local icon markers and identify whether there is an icon whose label name matches the initial icon marker's label name, in order to determine whether the initial icon marker and the local icon marker match. If the local icon markers are stored in a user-defined file repository, the file is opened using the fopen function, and then the local icon markers in the file are read using the fread function, and the local icon markers are searched for to see if there is an icon that matches the initial icon marker.

[0106] In this embodiment of the invention, by determining whether the business icon to be matched matches the local icon library and the remote icon library respectively, the business icon to be matched can be automatically matched with the current business icon and all business icons involved in the business update and upgrade process during the business upgrade process, without spending a lot of manpower on icon matching, thereby improving the accuracy of icon matching during the business update and upgrade process.

[0107] Furthermore, in this embodiment of the invention, the method for determining whether the initial icon mark matches the remote icon mark is the same as the method for determining whether the initial icon mark matches the local icon mark, and will not be described again here.

[0108] S204. When the business icon to be matched matches the icon library, the matching icon mark in the icon library is selected as the target icon mark of the business icon to be matched.

[0109] In this embodiment of the invention, when the business icon to be matched matches the icon library, it means that there is no need to add the same or duplicate icons to the business during the business update and upgrade process. By selecting the matching icon mark in the icon library as the target icon mark of the business icon to be matched, it is not necessary to add the same icon to the business during each business update and upgrade. That is, there is no need to compile and transmit a large number of duplicate files during the business update and upgrade process, which can improve the efficiency of business update and upgrade. Furthermore, it eliminates the need to spend a lot of manpower on icon matching, enabling automatic icon matching and further improving the accuracy of icon matching.

[0110] In this embodiment of the invention, the target icon marker refers to the current icon label required to match the business icon during the business update and upgrade process.

[0111] As an embodiment of the present invention, selecting a matching icon marker from the icon library as the target icon marker for the business icon to be matched includes:

[0112] Obtain the business requirements for the business icon to be matched, and the list of icons in the icon library that match the business icon to be matched;

[0113] Select the target icon marker of the business icon to be matched from the icon list according to the business requirements.

[0114] The business requirements refer to the functional issues that need to be addressed during the business update and upgrade process. For example, in the process of updating and upgrading insurance business, the business requirement could be to prioritize displaying the insurance icon with the highest purchase quantity.

[0115] In one embodiment of the present invention, the icon list refers to a data table that stores all similar icon markers that match the business icon to be matched, including various shapes and design styles. For example, if the business icon to be matched is a car insurance icon, then the icon list refers to a list of all car insurance icon markers.

[0116] For example, if there is a car insurance icon to be matched, and the business requirement is to prioritize displaying the car insurance icon with the highest purchase quantity, then the car insurance icon with the highest purchase quantity is selected from the list of all car insurance icon icons as the target icon for the business icon to be matched.

[0117] In an optional embodiment of the present invention, after determining the target icon mark, the system receives the user's click instruction on the target icon mark from the business front-end page. Based on the instruction, the system automatically imports the import path and import code of the target icon mark from the clipboard of the business front-end page. The system then directly pastes the target icon mark into the business front-end page component that needs to be imported based on the import path and import code, so as to realize the reference of the target icon mark in the business update and upgrade process.

[0118] S205. When the business icon to be matched does not match the icon library, an icon mark for the business icon to be matched is added to the business.

[0119] In this embodiment of the invention, when the business icon to be matched does not match the icon library, it means that there is no identical icon matching the business icon to be matched during the business update and upgrade process. By adding an icon marker for the business icon to be matched in the business, icons that did not appear during the business update and upgrade process can be added, further improving the icon library during the business update and upgrade process, and facilitating the improvement of the efficiency of business update and upgrade in the next business update and upgrade process.

[0120] As an embodiment of the present invention, adding the icon marker of the business icon to be matched in the business includes:

[0121] Identify the icon category of the business icon to be matched;

[0122] The business path of the business icon to be matched is determined based on the icon category;

[0123] The icon of the business to be matched is added to the business based on the business path.

[0124] The icon categories include functional icons (such as arrows and buttons), user icons (such as phone book icons and user avatar icons), and insurance business icons (such as car insurance icons and engineering insurance icons).

[0125] In one embodiment of the present invention, the business path corresponding to the business icon to be matched can be obtained through SQL query language based on the icon category, and the business to be matched can be added to the business category through SQL add language. For example, if the business icon to be matched is a left arrow, the business path corresponding to the left arrow can be found through SQL query language as the function class storage path, and the left arrow can be added to the function business category using SQL add language based on the function class storage path.

[0126] Compared with the prior art, the embodiments of this application have the following main advantages:

[0127] After obtaining the trained icon recognition model and business icon set, this application first uses the icon recognition model to recognize the icons in the business icon set, obtaining a business icon tag set. This allows for icon recognition of all business icons during business updates and upgrades, eliminating the need for secondary recognition of already recognized icons and improving icon recognition efficiency. Second, by storing the business icon tag set in a preset icon library, the same icons can be reused during business updates and upgrades, improving efficiency. Finally, by determining whether the business icon to be matched matches the icon library, the application solves the problem of expending significant manpower on icon matching during business updates and upgrades, thereby improving the accuracy of icon matching during business updates. When a match is found, the matching icon tag in the icon library is selected as the target icon tag for the business icon to be matched. This not only achieves icon deduplication, improving the efficiency of business updates and upgrades, but also enables automatic icon matching, further improving the accuracy of icon matching. When a mismatch is found, icon tags for the business icons to be matched are added to the business, which improves the icon library during icon matching and further enhances the accuracy of icon matching. Therefore, the icon matching method proposed in this application can improve the accuracy of icon matching and the efficiency of business updates and upgrades.

[0128] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0129] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware with computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When executed, the program can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0130] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0131] Further reference Figure 4 As a response to the above Figure 2 To implement the method shown, this application provides an embodiment of an icon matching device, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0132] like Figure 4 As shown, the icon matching device 400 described in this embodiment includes: a recognition module 401, a storage module 402, a matching module 403, a selection module 404, and a adding module 405. Wherein:

[0133] The recognition module 401 is used to acquire the trained icon recognition model and business icon set, and to use the icon recognition model to perform icon recognition on the icons in the business icon set to obtain a business icon tag set.

[0134] The storage module 402 is used to store the business icon mark set into a preset icon library;

[0135] The matching module 403 is used to obtain the business icon to be matched and determine whether the business icon to be matched matches the icon library;

[0136] The selection module 404 is used to select the matching icon mark in the icon library as the target icon mark of the business icon to be matched when the business icon to be matched matches the icon library.

[0137] The newly added module 405 is used to add an icon mark for the business icon to be matched in the business when the business icon to be matched does not match the icon library.

[0138] In this embodiment, the functions of each module / unit are as follows:

[0139] In detail, each module in the icon matching device 400 described in this embodiment of the invention employs the same methods as described above during use. Figures 1 to 3 The same technical means are used as the icon matching method described in the article, and can produce the same technical effect, so it will not be repeated here.

[0140] In this embodiment, after obtaining the trained icon recognition model and business icon set, the icon recognition model is first used to recognize the icons in the business icon set to obtain a business icon tag set. This allows for icon recognition of all business icons during business updates and upgrades, eliminating the need for secondary recognition of already recognized icons and improving icon recognition efficiency. Secondly, by storing the business icon tag set in a preset icon library, the same icons can be reused during business updates and upgrades, improving efficiency. Finally, by determining whether the business icon to be matched matches the icon library, the problem of requiring significant manpower for icon matching during business updates and upgrades can be solved, thereby improving the accuracy of icon matching during business updates. When a match is found, the matching icon tag in the icon library is selected as the target icon tag for the business icon to be matched. This not only achieves icon deduplication and improves the efficiency of business updates and upgrades but also enables automatic icon matching, further improving the accuracy of icon matching. When a mismatch is found, an icon tag for the business icon to be matched is added to the business. This improves the icon library during icon matching, further enhancing the accuracy of icon matching. Therefore, the icon matching device proposed in this application can improve the accuracy of icon matching and the efficiency of business updates and upgrades.

[0141] The computer device 5 includes a memory 51, a processor 52, and a network interface 53 that are interconnected via a system bus. It should be noted that only the computer device 5 with components 51-53 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0142] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0143] The memory 51 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or D icon matching memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 51 may be an internal storage unit of the computer device 5, such as the hard disk or memory of the computer device 5. In other embodiments, the memory 51 may also be an external storage device of the computer device 5, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 5. Of course, the memory 51 may include both the internal storage unit and its external storage device of the computer device 5. In this embodiment, the memory 51 is typically used to store the operating system and various application software installed on the computer device 5, such as computer-readable instructions for icon matching methods. In addition, the memory 51 can also be used to temporarily store various types of data that have been output or will be output.

[0144] In some embodiments, the processor 52 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 52 is typically used to control the overall operation of the computer device 5. In this embodiment, the processor 52 is used to execute computer-readable instructions stored in the memory 51 or to process data, for example, to execute computer-readable instructions for the icon matching method.

[0145] The network interface 53 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 5 and other electronic devices.

[0146] In this embodiment, after obtaining the trained icon recognition model and business icon set, the icon recognition model is first used to recognize the icons in the business icon set to obtain a business icon tag set. This allows for icon recognition of all business icons during business updates and upgrades, eliminating the need for secondary recognition of already recognized icons and improving icon recognition efficiency. Secondly, by storing the business icon tag set in a preset icon library, the same icons can be reused during business updates and upgrades, improving efficiency. Finally, by determining whether the business icon to be matched matches the icon library, the problem of requiring significant manpower for icon matching during business updates and upgrades can be solved, thereby improving the accuracy of icon matching during business updates. When a match is found, the matching icon tag in the icon library is selected as the target icon tag for the business icon to be matched. This not only achieves icon deduplication and improves the efficiency of business updates and upgrades but also enables automatic icon matching, further improving the accuracy of icon matching. When a mismatch is found, an icon tag for the business icon to be matched is added to the business. This improves the icon library during icon matching, further enhancing the accuracy of icon matching. Therefore, the icon matching computer device proposed in this application can improve the accuracy of icon matching and the efficiency of business updates and upgrades.

[0147] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the icon matching method described above.

[0148] In this embodiment, after obtaining the trained icon recognition model and business icon set, the icon recognition model is first used to recognize the icons in the business icon set to obtain a business icon tag set. This allows for icon recognition of all business icons during business updates and upgrades, eliminating the need for secondary recognition of already recognized icons and improving icon recognition efficiency. Secondly, by storing the business icon tag set in a preset icon library, the same icons can be reused during business updates and upgrades, improving efficiency. Finally, by determining whether the business icon to be matched matches the icon library, the problem of requiring significant manpower for icon matching during business updates and upgrades can be solved, thereby improving the accuracy of icon matching during business updates. When a match is found, the matching icon tag in the icon library is selected as the target icon tag for the business icon to be matched. This not only achieves icon deduplication and improves the efficiency of business updates and upgrades but also enables automatic icon matching, further improving the accuracy of icon matching. When a mismatch is found, an icon tag for the business icon to be matched is added to the business. This improves the icon library during icon matching, further enhancing the accuracy of icon matching. Therefore, the icon matching storage medium proposed in this application can improve the accuracy of icon matching and the efficiency of business updates and upgrades.

[0149] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0150] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. An icon matching method, characterized by, The method comprises the following steps: obtaining a trained icon recognition model and a business icon set, performing icon recognition on icons in the business icon set by using the icon recognition model, and obtaining a business icon label set; storing the business icon label set in a preset icon library; obtaining a to-be-matched business icon, and determining whether the to-be-matched business icon matches the icon library; when the to-be-matched business icon matches the icon library, selecting a matching icon label in the icon library as a target icon label of the to-be-matched business icon; when the to-be-matched business icon does not match the icon library, adding an icon label of the to-be-matched business icon in a business.

2. The icon matching method of claim 1, wherein, The determination of whether the to-be-matched business icon matches the icon library comprises: obtaining an initial icon label of the to-be-matched business icon, a local icon label in the icon library, and a remote icon label in the icon library; determining whether the initial icon label matches the local icon label; when the initial icon label matches the local icon label, it is determined that the to-be-matched business icon matches the icon library; when the initial icon label does not match the local icon label, it is determined whether the initial icon label matches the remote icon label; when the initial icon label matches the remote icon label, it is determined that the to-be-matched business icon matches the icon library; when the initial icon label does not match the remote icon label, it is determined that the to-be-matched business icon does not match the icon library.

3. The icon matching method of claim 1, wherein, The selection of the matching icon label in the icon library as the target icon label of the to-be-matched business icon comprises: obtaining a business requirement of the to-be-matched business icon, and an icon list matching the to-be-matched business icon in the icon library; selecting a target icon label of the to-be-matched business icon from the icon list according to the business requirement.

4. The icon matching method of claim 1, wherein, The addition of the icon label of the to-be-matched business icon in the business comprises: identifying an icon category of the to-be-matched business icon; determining a business path of the to-be-matched business icon according to the icon category; adding the to-be-matched business icon to the business based on the business path.

5. The icon matching method according to any one of claims 1 to 4, characterized in that, The storage of the business icon label set in the preset icon library comprises: obtaining an icon path of the business icon label set, and identifying a business category to which an icon in the business icon label set belongs; storing the business icon label set in a corresponding icon library according to the icon path and the business category.

6. The icon matching method according to any one of claims 1 to 4, wherein, The icon recognition on the icons in the business icon set by using the icon recognition model to obtain the business icon label set comprises: performing convolution operation on the icons in the business icon set by using a convolution layer of the icon recognition model to obtain an icon feature set; performing maximum pooling operation on the icon feature set by using a pooling layer of the icon recognition model to obtain a pooled icon set; inputting the pooled icon set into an activation function in a full connection layer in the icon recognition model, and outputting the business icon label set.

7. The icon matching method according to any one of claims 1 to 4, wherein, Before the obtaining of the trained icon recognition model and the business icon set, the method further comprises: Obtain the training icon set and the preset model training framework; Configure the icon modeling workflow of the training icon set in the visualization modeling module of the model training framework, and create an initial icon recognition model according to the icon modeling workflow; Obtain the real icon label set corresponding to the training icon set, input the training icon set into the initial icon recognition model, and obtain the predicted icon label set of the training icon set; Calculate the loss value between the predicted icon label set and the real icon label set, adjust the parameters of the initial icon recognition model according to the loss value, until the loss value meets a preset threshold, and obtain the trained icon recognition model.

8. An icon matching apparatus characterized by comprising: include: The recognition module is used to acquire the trained icon recognition model and business icon set, and to use the icon recognition model to recognize the icons in the business icon set to obtain a business icon tag set. A storage module is used to store the business icon mark set into a preset icon library; The matching module is used to obtain the business icon to be matched and determine whether the business icon to be matched matches the icon library; The selection module is used to select the matching icon mark in the icon library as the target icon mark of the business icon to be matched when the business icon to be matched matches the icon library. A new module is added to the business if the icon to be matched does not match the icon library, and then adds an icon tag for the icon to be matched to the business.

9. A computer device comprising a memory and a processor, the memory storing computer-readable instructions, wherein the processor, when executing the computer-readable instructions, implements the steps of the icon matching method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the icon matching method as described in any one of claims 1 to 7.

Images