Graphic design material classification management system with user-defined retrieval

Abstract

The invention relates to a custom retrieval graphic design material classification management system, which relates to the field of information retrieval, and is characterized in that a user side receives a request containing a material index target character, and a material character grading module of a server side grades the request to obtain N-grade index target characters. And the material index module retrieves the (k-1)-level index set according to the k-level index target character to obtain a k-level index set, and the initial k = 1, 0-level index set is a plane material library. A loop judgment module judges whether the k-level set reaches a material quantity threshold value or not, and a loop decision module decides klt; if N is greater than N, returning to the index module to continue circulation, otherwise, sending the k-level index set to the user side. According to the method and the device, the technical problem of low efficiency and accuracy of material retrieval by a user due to a classified storage mode of an existing plane material library is solved, and the technical effect of improving the efficiency and accuracy of plane design material retrieval by the user is achieved through modes of user-defined retrieval and multi-level indexing.

Description

Technical Field

[0001] The present invention relates to the field of information retrieval, and in particular to a graphic design material classification management system for customized retrieval. Background Art

[0002] With the development of the graphic design industry, the demand for design materials is also increasing. In order to facilitate management and use, the existing technology generally adopts the method of classifying and storing according to material attributes to build a graphic material library. Common material classifications include vector materials, layered materials, design drawings, PSD, wallpapers, high-definition pictures, patterns, PNG, fonts, etc. However, this traditional classification storage method has certain limitations. When users need to find specific materials, they often need to browse and filter a lot in the corresponding material types to find suitable materials. This process is not only time-consuming and labor-intensive, but also has low retrieval efficiency and accuracy, which cannot meet the user's needs to quickly and accurately obtain the required materials. In addition, with the continuous expansion of the scale of the material library, the traditional classification method also faces challenges in scalability and maintainability. When the number of materials reaches a certain scale, it becomes increasingly difficult to rely solely on manual classification and management, and it is easy to have problems such as inaccurate classification and material duplication, which further affects the efficiency and experience of user retrieval of materials. Summary of the invention

[0003] The present invention aims to solve the technical problem that the classification storage method of the existing graphic material library leads to low efficiency and accuracy in user material retrieval, and provides a graphic design material classification management system with customized retrieval.

[0004] The technical solution of the present invention to solve the above technical problems is as follows: In a first aspect, the present invention provides a custom-retrieval graphic design material classification management system, comprising: a user end, for receiving a material index request, wherein the material index request includes a material index target trait; a server end, the server end comprising: a material trait grading module, for grading the material index target trait, obtaining a first-level material index target trait, a second-level material index target trait, and up to an N-level material index target trait; a material indexing module, for retrieving a k-1-level index material set according to a k-level material index target trait, and obtaining a k-level index material set, wherein the initial value of k is equal to 1, and the 0-level index material set is a graphic material library; a loop judgment module, for judging whether the k-level index material set is greater than or equal to a material quantity threshold; a loop decision module, for when the k-level index material set is greater than or equal to the material quantity threshold, and k<N, k is added by one, and the k-level index material set is returned to the material indexing module to execute a loop, and when the k-level index material set is less than the material quantity threshold, or / and k=N, the k-level index material set is sent to the user end.

[0005] In a second aspect, the present invention provides a method for classifying and managing graphic design materials with customized retrieval, comprising: receiving a material index request, wherein the material index request includes a material index target trait; grading the material index target trait to obtain a first-level material index target trait, a second-level material index target trait, and finally an N-level material index target trait; searching a k-1-level index material set according to the k-level material index target trait to obtain a k-level index material set, wherein the initial value of k is equal to 1, and the 0-level index material set is a graphic material library; determining whether the k-level index material set is greater than or equal to a material quantity threshold; when the k-level index material set is greater than or equal to the material quantity threshold, and k<N, k is increased by one, and the k-level index material set is returned to the material index module to execute a loop, and when the k-level index material set is less than the material quantity threshold, or / and k=N, the k-level index material set is sent to the user terminal.

[0006] In a third aspect, the present application provides an electronic device, comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is used to execute a custom retrieval graphic design material classification management system provided by the present application.

[0007] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program for executing a custom retrieval graphic design material classification management system provided by the present application.

[0008] The beneficial effects of the present invention are: First, the system sets up a user end and a server end. The user end is used to receive a material index request input by a user, and the request includes the material target properties that the user expects to retrieve. By introducing user-defined index target properties, the system can provide personalized material retrieval services to meet the user's specific material needs. On the server end, a material property grading module is set. This module performs grading processing on the material index target properties input by the user, and divides them into first-level index target properties, second-level index target properties, and finally N-level index target properties. By grading the target properties, more fine-grained material retrieval can be achieved and the accuracy of retrieval can be improved. Next, the material index module searches the material library step by step according to the graded target properties. Specifically, the material set obtained by the k-1 level index is searched according to the k-level index target properties to obtain the k-level index material set. Among them, the initial value of k is 1, and the 0-level index material set is the initial plane material library. Through step-by-step retrieval, the retrieval range can be continuously narrowed and the retrieval efficiency can be improved. In the process of step-by-step retrieval, the cyclic discrimination module determines whether the material set obtained by each level index reaches the preset material quantity threshold. If the threshold is reached and the maximum number of levels N is not reached, the next level of indexing will be entered; if the threshold is not reached or the maximum number of levels is reached, the index material set of the current level will be returned to the user end as the final search result. By introducing the material quantity threshold and the maximum number of levels, a balance is achieved between retrieval accuracy and retrieval efficiency. After that, the user end receives the k-level index material set returned by the server, and can obtain the graphic design materials that meet the custom index target characteristics.

[0009] In summary, by building a graphic design material classification management system with customized retrieval, the problems of low retrieval efficiency and accuracy caused by the traditional graphic material library classification storage method are overcome, and personalized, efficient and accurate material retrieval services are realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the structure of a graphic design material classification management system with customized retrieval provided by the present invention; Figure 2 A schematic diagram of a process flow of a graphic design material classification management method for custom retrieval provided by the present invention; Figure 3 A schematic diagram of the structure of an electronic device provided by the present invention; Figure 4 A schematic diagram of the structure of a computer-readable storage medium provided by the present invention.

[0011] In the accompanying drawings, the components represented by the reference numerals are as follows: The user end 110 , the server end 120 , the material property classification module 121 , the material indexing module 122 , the loop identification module 123 , the loop decision module 124 , the electronic device 200 , the memory 210 , the processor 220 , the first computer program 211 , the computer-readable storage medium 300 , and the second computer program 311 . DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0013] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0014] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0015] Embodiment 1: like Figure 1 As shown, the embodiment of the present invention provides a graphic design material classification management system for customized retrieval, including: The client 110 is configured to receive a material index request, wherein the material index request includes a material index target property.

[0016] Specifically, in the graphic design material classification management system of customized retrieval, the user terminal 110 is responsible for receiving the material index request from the user. When the user needs to retrieve graphic design materials of specific characteristics, the user enters the specific retrieval conditions, i.e., the material index target characteristics, through the human-computer interaction interface provided by the user terminal 110. For example, the user can enter the target characteristics such as "red", "vector", "flower", etc., which are used to describe the characteristic information such as color, material type, content theme, etc. that the required material should have.

[0017] After receiving the index request input by the user, the client 110 will encapsulate it into a standardized material index request, which includes one or more material index target properties input by the user. Afterwards, the client 110 will send the encapsulated material index request to the server 120 via network transmission, and the server 120 will perform subsequent index processing.

[0018] By receiving the target properties of the material index through the user terminal 110, the user can flexibly and accurately describe the property characteristics of the target material according to his actual needs, thereby realizing a personalized and customized material retrieval function. Compared with the traditional method of organizing materials according to a fixed classification system, the indexing method based on target properties can significantly improve the flexibility and practicality of material retrieval, allowing users to find design materials that meet their needs more quickly and accurately, thereby improving the work efficiency of graphic design.

[0019] In the graphic design material classification management system of custom retrieval, in addition to the user terminal 110, it also includes a server terminal 120 that is connected to the communication terminal. The server terminal 120 is mainly responsible for processing the material index request sent by the user terminal 110, and performing the related functions of material retrieval and management. The server terminal 120 includes a material property grading module 121, a material indexing module 122, a loop discrimination module 123 and a loop decision module 124, which are respectively used to complete different processing stages and tasks of the material index request. These functional modules jointly realize the custom retrieval and classification management of graphic design materials through mutual cooperation and data interaction. The specific implementation method of each module will be described in detail below.

[0020] The material property classification module 121 is used to classify the material index target properties to obtain the first-level material index target properties, the second-level material index target properties and the N-level material index target properties.

[0021] Specifically, the material property classification module 121 first receives a material index request from the user terminal 110, and extracts one or more material index target properties from it. Then, the module analyzes and evaluates these target properties one by one, and divides them into different importance levels to obtain the first-level material index target properties, the second-level material index target properties, the third-level material index target properties, and so on, until the N-level material index target properties. Among them, N is a positive integer that can be set according to actual application requirements, indicating the maximum number of grades of the index target properties.

[0022] Different levels of index target traits have different importance and priority. Generally speaking, the first-level index target traits represent the material attribute characteristics that users pay the most attention to and value the most, and should be given priority and satisfied in subsequent material retrieval and screening; while the second-level, third-level and even N-level index target traits are relatively unimportant and can be further refined and supplemented on the basis of the first-level material index target traits.

[0023] By introducing multi-level index target properties, the material property classification module 121 can effectively express and organize the user's search intention in a hierarchical manner, making the subsequent material retrieval and management process more intelligent. At the same time, this hierarchical processing mechanism can also improve the operating efficiency and response speed of the system, reduce unnecessary computing overhead, and thus significantly improve the actual user experience.

[0024] The material indexing module 122 is used to retrieve the k-1 level index material set according to the k level material index target property to obtain the k level index material set, where the initial value of k is 1, and the 0 level index material set is a flat material library.

[0025] Specifically, the material indexing module 122 first receives the primary material index target properties, the secondary material index target properties, and so on, output by the material property classification module 121, until the N-level material index target properties. Then, the module initializes an index level variable k and sets its value to 1, indicating that the primary index target properties are currently being retrieved.

[0026] During the retrieval process, the material index module 122 matches and screens the index material set of the previous level (i.e., level k-1) one by one according to the current k-level material index target trait. Among them, when k is equal to 1, the k-1-level index material set is actually the entire plane material library, also known as the 0-level index material set. The material index module 122 traverses each material object in the k-1-level index material set, extracts its related attribute feature values, and compares them with the k-level material index target trait. If the attribute feature value of a material object has a high degree of match with the k-level index target trait, it is added to the k-level index material set; otherwise, it is removed. After a complete round of traversal and screening, the material index module 122 outputs the obtained k-level index material set as the basis for the next level (i.e., level k+1) material index. Through the level-by-level retrieval and screening method, the material index module 122 continuously narrows the scope of candidate materials until the target material set that ultimately meets the user's needs is obtained.

[0027] The loop determination module 123 is used to determine whether the k-level index material set is greater than or equal to a material quantity threshold.

[0028] Specifically, the loop identification module 123 receives the k-level index material set output by the material index module 122 and obtains a preset material quantity threshold, which represents the minimum material quantity that the system considers sufficient to meet user needs and can be dynamically adjusted according to actual application scenarios and user feedback.

[0029] The loop discrimination module 123 calculates the number of material objects contained in the current k-level index material set, and compares it with the material quantity threshold value. If the number of the k-level index material set is greater than or equal to the material quantity threshold value, it means that the current search result is rich and comprehensive enough to meet the actual needs of the user. At this time, the loop discrimination module 123 sends an affirmative discrimination result to the loop decision module 124, indicating that the current search process can be ended. On the contrary, if the number of the k-level index material set is less than the material quantity threshold value, it means that the current search result is not ideal enough and needs to be further refined and deepened. At this time, the loop discrimination module 123 sends a negative discrimination result to the loop decision module 124, indicating that it is necessary to continue to perform the index target trait retrieval of the next level.

[0030] By introducing the loop discrimination module 123, the quality and quantity of the current results are dynamically evaluated during the retrieval process to avoid the problem of over-retrieval or under-retrieval, thereby reducing unnecessary computing overhead as much as possible while ensuring the relevance of the results, thereby improving the overall efficiency of the system. At the same time, the loop discrimination module 123 can also cooperate with other modules to implement intelligent optimization strategies. For example, the material quantity threshold is dynamically adjusted according to the user's historical behavior and feedback, or the quantity and quality of subsequent results are predicted in real time during the retrieval process, so as to more accurately control the depth and breadth of the retrieval.

[0031] The loop decision module 124 is used to, when the k-level index material set is greater than or equal to the material quantity threshold and k<N, add one to k and return the k-level index material set to the material index module to execute the loop; when the k-level index material set is less than the material quantity threshold, or / and k=N, send the k-level index material set to the user terminal.

[0032] Specifically, the loop decision module 124 receives the discrimination result sent by the loop discrimination module 123, and performs comprehensive analysis and decision-making in combination with the current index level variable k and the maximum number of classifications N.

[0033] In one case, if the judgment result of the loop judgment module 123 indicates that the number of the current k-level index material set is greater than or equal to the material number threshold, and the index level k at this time is still less than the maximum number of levels N, it means that the current search result is already relatively ideal, but has not yet reached the finest granularity. At this time, the loop decision module 124 adds 1 to the value of the index level variable k, indicating that it is necessary to enter the next level of search. At the same time, the module resends the current k-level index material set to the material index module 122 as the input of a new round of search to further improve the accuracy of the results.

[0034] In another case, if the judgment result of the loop judgment module 123 indicates that the number of the current k-level index material set is still less than the material number threshold, or the current index level k is equal to the maximum level number N, it means that it is meaningless to continue the subdivision search, or the finest granularity has been reached. At this time, the loop decision module 124 directly sends the current k-level index material set to the user terminal 110, and presents it to the user as the final search result.

[0035] By introducing the loop decision module 124, the direction and depth of the search process are dynamically controlled according to the actual search situation, while satisfying the user's needs as much as possible, avoiding unnecessary waste of computing resources and improving the efficiency and accuracy of the user's search for graphic design materials.

[0036] Furthermore, the material property classification module 121 is used to classify the material index target property to obtain the first-level material index target property, the second-level material index target property, and finally the N-level material index target property, including: Obtain a trait grading table; According to the trait grading table, traverse the material index target trait to obtain the target trait level identifier; According to the target trait level identifier, the material index target traits are grouped to obtain a plurality of groups of material index target traits, wherein the plurality of groups of material index target traits have a plurality of target trait level identifiers; According to the target property level identifiers, the groups of material index target properties are sorted from small to large levels to obtain the first-level material index target property, the second-level material index target property, and finally the N-level material index target property.

[0037] In a feasible implementation, the material property classification module 121 adopts the following specific steps and methods in the process of classifying the material index target properties to obtain the first-level material index target properties, the second-level material index target properties and finally the N-level material index target properties.

[0038] First, the material property grading module 121 obtains a pre-constructed property grading table. The property grading table is a mapping relationship table obtained by the system after classifying and rating various common material properties according to the attribute characteristics of the graphic design materials and the user's usage habits. By referring to the property grading table, the material property grading module 121 can quickly determine the importance level of each material index target property. Then, the material property grading module 121 begins to traverse all material index target properties contained in the material index request sent by the user end 110. For each target property, the material property grading module 121 searches and matches in the property grading table to obtain the level identifier corresponding to the target property. The level identifier is a numerical value or symbol indicating the importance of the target property, which is divided into several different levels from low to high.

[0039] After completing the traversal of all target traits and obtaining the level identifiers, the material trait classification module 121 groups the material index target traits according to these level identifiers. Target traits with the same level identifier will be classified into the same group, and finally several groups of material index target traits of different importance will be obtained. Each group of target traits corresponds to a specific level identifier. After that, the material trait classification module 121 sorts several groups of material index target traits according to the size of the level identifier. The target trait group with a smaller level identifier is considered to have a higher importance and priority, so it will be ranked in front; conversely, the target trait group with a larger level identifier is considered to have a relatively low importance and will be ranked at the back. After this step of sorting processing, the material trait classification module 121 finally outputs an ordered target trait classification result, including a first-level material index target trait, a second-level material index target trait, a third-level material index target trait, and so on, until the N-level material index target trait.

[0040] The material property classification module 121 efficiently and accurately completes the classification processing of the material index target properties input by the user, providing an important basis and reference for subsequent graded retrieval. This automatic classification mechanism based on the property classification table can improve the system's understanding and expression of the user's search intention, thereby achieving a more accurate and humanized material recommendation service. At the same time, the mechanism also has strong flexibility and scalability, and can adapt to the ever-changing user needs and material characteristics by dynamically updating and optimizing the property classification table.

[0041] Furthermore, the construction process of the trait grading table is as follows: When the plane material library is updated, the first property trigger frequency of the plane material library, the second property trigger frequency of the plane material library, and so on until the Mth property trigger frequency of the plane material library are obtained; Based on the trait trigger frequency deviation threshold, the material traits are clustered based on the first trait trigger frequency of the plane material library, the second trait trigger frequency of the plane material library, and up to the Mth trait trigger frequency of the plane material library to obtain a first group of trait sets, a second group of trait sets, and up to the Qth group of trait sets; Traversing the first set of trait sets, the second set of trait sets, and up to the Qth set of trait sets to perform trigger frequency mean calculation, and obtaining a first trigger frequency mean, a second trigger frequency mean, and up to the Qth trigger frequency mean; According to the first trigger frequency mean, the second trigger frequency mean, up to the Qth trigger frequency mean, the first group of trait sets, the second group of trait sets, up to the Qth group of trait sets are sorted from small to large to obtain a first level identification trait set, a second level identification trait set, up to the Lth level identification trait set; The trait grading table is constructed based on the first level identification trait set, the second level identification trait set, and up to the Lth level identification trait set.

[0042] In a preferred embodiment, an automated construction process is provided for the property grading table used by the material property grading module 121, so as to dynamically generate and update the table according to the actual situation of the plane material library, thereby improving the timeliness and accuracy of material grading.

[0043] First, when the plane material library is updated, the system automatically triggers the construction process of the trait classification table. The update can be the addition, deletion or modification of material objects, or the change of material attributes, etc. Then, the trigger frequency of each material trait in the plane material library is counted, that is, the number of times each trait appears or is used in the material library, and the trigger frequency of the first trait of the plane material library, the trigger frequency of the second trait of the plane material library, and so on, are obtained, until the trigger frequency of the Mth trait of the plane material library. Among them, M represents the total number of different traits in the material library. Next, based on the preset trait trigger frequency deviation threshold, the trigger frequencies of each trait obtained in the previous step are clustered. Traits with similar trigger frequencies will be classified into the same group, thereby obtaining the first group of trait sets, the second group of trait sets, and so on, until the Qth group of trait sets. Among them, Q represents the total number of trait sets obtained by cluster analysis.

[0044] For each set of traits, all traits in it are traversed, and the corresponding trigger frequency means are calculated to obtain the first trigger frequency mean, the second trigger frequency mean, and so on, until the Qth trigger frequency mean. These means reflect the average trigger frequency levels of different trait groups. According to the obtained trigger frequency means, the Q trait sets are sorted. Trait sets with smaller trigger frequency means are considered to be more important and more commonly used, so they will be ranked in front; conversely, trait sets with larger trigger frequency means are considered to be relatively minor and less used, and will be ranked in the back. After this step of sorting, the first-level identification trait set, the second-level identification trait set, and so on are obtained, until the Lth-level identification trait set. Among them, L represents the total number of levels of trait grading. Afterwards, a complete trait grading table is constructed based on the trait sets of each level identification obtained in the previous step. Each row of the table corresponds to a trait grade, including the identification of the level and the specific traits contained in the level.

[0045] When sorting the trait sets, sets with lower trigger frequencies will be prioritized. This is because traits with lower trigger frequencies usually represent some more unique and detailed attributes, which are very helpful for quickly narrowing the scope of material retrieval and improving retrieval accuracy. On the contrary, traits with higher trigger frequencies may be too broad and general, and the retrieval efficiency is relatively low. In addition, in the process of constructing the trait grading table, each trait is actually defined and described by multiple attribute characteristic values. Therefore, when statistically analyzing the trait trigger frequency and performing cluster analysis, it is necessary to comprehensively consider the situation of each attribute characteristic value to ensure the rationality and accuracy of the grading results.

[0046] Through the above construction process, the property classification table is dynamically generated and updated according to the real-time status of the plane material library, thereby providing a highly adaptive reference basis for the material property classification module 121. This mechanism can effectively improve the efficiency and accuracy of material classification and retrieval, and also has strong flexibility and robustness, and can adapt to the frequent changes and updates of the plane material library.

[0047] Further, when the plane material library is updated, obtaining the first property trigger frequency of the plane material library, the second property trigger frequency of the plane material library, and up to the Mth property trigger frequency of the plane material library includes: Obtaining a first attribute characteristic value set; Based on the trait characteristic value deviation threshold, cluster analysis is performed on the first trait attribute characteristic value set to obtain a trait attribute characteristic value clustering result; Any one of the clustering results of the trait attribute characteristic values ​​belongs to the first trait of the plane material library, the second trait of the plane material library, and up to the Mth trait of the plane material library, and the number of characteristic values ​​of any one of the clustering results of the trait attribute characteristic values ​​belongs to the first trait triggering frequency of the plane material library, the second trait triggering frequency of the plane material library, and up to the Mth trait triggering frequency of the plane material library; After each property attribute characteristic value set is analyzed, the first property trigger frequency of the plane material library, the second property trigger frequency of the plane material library, and so on until the Mth property trigger frequency of the plane material library are output.

[0048] In a preferred embodiment, when the plane material library is updated, first, the system obtains the attribute feature value set corresponding to each material property in the plane material library. The material property is defined and described by multiple attribute feature values, such as color, shape, size, style, etc. Each attribute feature value set contains all the specific values ​​of the property appearing in the material library. Then, for each attribute feature value set, cluster analysis is performed based on a preset property feature value deviation threshold. The property feature value deviation threshold represents the similarity between the attribute feature values, and the feature values ​​with a deviation less than the threshold value will be classified into the same class. After cluster analysis, a property attribute feature value clustering result is obtained, which contains several classes, and each class is composed of a group of similar attribute feature values. Next, each class in the property attribute feature value clustering result is mapped to a specific material property, that is, the first property of the plane material library, the second property of the plane material library, and so on, until the Mth property of the plane material library. The purpose of this step is to recombine the scattered attribute feature values ​​into a complete material property so as to perform subsequent statistics on the trigger frequency.

[0049] At the same time, the number of attribute characteristic values ​​contained in each class is used as the trigger frequency of the material trait corresponding to the class, that is, the trigger frequency of the first trait of the plane material library, the trigger frequency of the second trait of the plane material library, and so on, until the trigger frequency of the Mth trait of the plane material library. This is because the number of times an attribute characteristic value appears actually reflects the frequency with which the material trait containing the characteristic value is used or triggered. Afterwards, when each set of trait attribute characteristic values ​​in the plane material library has completed the above-mentioned clustering analysis and mapping, the final trigger frequency of the first trait of the plane material library, the trigger frequency of the second trait of the plane material library, and so on, are output as the intermediate result of the trait grading table construction process, and are used for subsequent steps such as trait clustering and sorting.

[0050] Through the trigger frequency statistics method based on attribute feature value clustering, the actual usage of each material trait in the plane material library can be reflected more accurately and comprehensively, and the statistical errors caused by improper trait definition or scattered attribute values ​​can be reduced, thereby providing a more reliable data basis for the automatic construction of trait grading tables.

[0051] Furthermore, the embodiment of the present application also includes: According to the k-1 level index material set, obtaining a k-1 level first index material; Extracting the k-1 level first index material having a k level material property feature value set; Receive k-level material index target trait deviation threshold; Counting the proportion of the number of trait attributes of the k-level material index target trait and the k-level material trait feature value set that is less than or equal to the k-level material index target trait deviation threshold, and setting it as k-level proximity; When the k-level proximity is greater than or equal to a proximity threshold, the k-1-level first index material is added to the k-level index material set.

[0052] In a preferred embodiment, in the process of obtaining a k-level index material set, a retrieval method based on a property feature value deviation and a proximity threshold is adopted to improve the accuracy and efficiency of material retrieval.

[0053] First, the material index module 122 obtains a material object to be retrieved from the k-1-level index material set, which is called the k-1-level first index material. The k-1-level index material set here refers to the material subset obtained after the first k-1-level material index target property screening. When k is equal to 1, the k-1-level index material set is actually the entire plane material library. Then, the material index module 122 extracts all the k-level material property feature values ​​possessed by the k-1-level first index material to form a k-level material property feature value set. The k-level material property feature value refers to a more specific and detailed attribute value corresponding to the k-level material index target property. For example, when the k-level material index target property is "red", the corresponding k-level material property feature value may include "dark red", "light red", "pink", etc. Next, the material index module 122 receives a preset k-level material index target property deviation threshold. The threshold represents the maximum deviation degree allowed between the k-level material index target property and the k-level material property feature value, which is used to measure the similarity or matching degree between the two.

[0054] For each feature value in the set of k-level material properties feature values ​​of the k-1-level first index material, the material index module 122 calculates the deviation between it and the k-level material index target property, and determines whether the deviation is less than or equal to the preset deviation threshold. If yes, the feature value is marked as a match; otherwise, it is marked as a mismatch. After completing the matching judgment of all k-level material property feature values, the material index module 122 will count the proportion of the number of matching feature values ​​to the total number of feature values, and define it as the k-level proximity. The k-level proximity reflects the overall similarity between the k-1-level first index material and the k-level material index target property, and the value range is between 0 and 1. Afterwards, the material index module 122 compares the k-level proximity with the preset proximity threshold. If the k-level proximity is greater than or equal to the proximity threshold, it is considered that the k-1-level first index material is close enough to the k-level material index target property to meet the user's retrieval needs, so it is added to the k-level index material set; otherwise, it is considered that the two are very different and are not adopted. The material indexing module 122 repeatedly executes the above retrieval steps for each material object in the k-1 level index material set, and finally obtains a complete k-level index material set as the k-th level index result output.

[0055] Through the above-mentioned retrieval method based on trait feature value deviation and proximity threshold, the material indexing module 122 can flexibly control the granularity and accuracy of material screening in each level of indexing process, which not only ensures the relevance of the retrieval results to user needs, but also avoids the problem of too few or empty retrieval results caused by overly strict matching conditions, thereby significantly improving the practicality of material retrieval and user experience.

[0056] Furthermore, the proximity threshold is equal to 1.

[0057] Specifically, the proximity threshold is a parameter used to measure the overall similarity between the k-1 level index material and the target trait of the k level material index, and its value range is between 0 and 1. The larger the proximity threshold, the higher the requirement for material similarity, the higher the accuracy of the retrieval results, but the recall rate may be reduced; conversely, the smaller the proximity threshold, the looser the requirement for material similarity, the lower the accuracy of the retrieval results, but the recall rate will be improved.

[0058] When the proximity threshold is equal to 1, it means that in each level of retrieval process, the material indexing module 122 will only add the material to the k-level index material set when the k-1 level index material's k-level material property feature value set completely matches the k-level material index target property. In other words, the material retrieval in this case adopts a very strict matching strategy, requiring that there should be no attribute differences or deviations between the candidate material and the target property. This strict matching strategy can ensure the high accuracy and relevance of the retrieval results, ensuring that the retrieved materials fully meet the index target properties at all levels specified by the user, thereby meeting the actual needs of the user to the greatest extent and providing the most accurate material retrieval. Especially in some application scenarios that have extremely high requirements for material quality and consistency, such as corporate VI design, product appearance design, etc., the use of a retrieval strategy with a proximity threshold equal to 1 can well control the standardization and uniformity of the material, and avoid problems such as confusion in design style caused by the use of improper materials.

[0059] The embodiment of the present invention provides a custom search graphic design material classification management system, which has at least the following technical effects: The user end is used to receive a material index request, wherein the material index request includes a material index target trait, allowing the user to input a customized material index target trait, realizing personalized material retrieval service, and meeting the user's specific material needs. The material trait grading module is used to grade the material index target trait, obtain the first-level material index target trait, the second-level material index target trait, and finally the N-level material index target trait, and divide the material index target trait input by the user into multiple levels to realize more fine-grained material retrieval and improve the accuracy of retrieval. The material index module is used to retrieve the k-1-level index material set according to the k-level material index target trait, and obtain the k-level index material set. The initial value of k is equal to 1, and the 0-level index material set is a flat material library. Through step-by-step retrieval, the retrieval scope is continuously narrowed to improve the retrieval efficiency. Each level of retrieval is based on the material set obtained by the previous level index, and gradually approaches the final target material. The cyclic discrimination module is used to determine whether the k-level index material set is greater than or equal to the material quantity threshold. By introducing the material quantity threshold, the scale of the material set obtained by each level of index is controlled. When the size of the set reaches the threshold, it indicates that enough candidate materials have been obtained to enter the next level of indexing or return results. The loop decision module is used to add one to k when the k-level index material set is greater than or equal to the material quantity threshold, and k < N, and return the k-level index material set to the material index module to execute the loop. When the k-level index material set is less than the material quantity threshold, or / and k=N, the k-level index material set is sent to the user end, and the next behavior of the system is determined based on the judgment result of the loop judgment module. If the index material set of the current level does not reach the threshold and does not reach the maximum number of levels, enter the next level of indexing; otherwise, return the current set to the user end as the final result.

[0060] In summary, the system can perform multi-level indexing and retrieval of the plane material library according to the index target characteristics customized by the user, and finally provide the user with highly matched material results. This customized retrieval method overcomes the limitations of the traditional classification storage method of the plane material library and significantly improves the efficiency and accuracy of material retrieval.

[0061] Embodiment 2: like Figure 2 As shown, based on the same inventive concept of a graphic design material classification management system for custom retrieval provided in Embodiment 1, an embodiment of the present invention further provides a graphic design material classification management method for custom retrieval, comprising: receiving a material index request, wherein the material index request includes a material index target trait; The material index target traits are graded to obtain a first-level material index target trait, a second-level material index target trait, and finally an N-level material index target trait; According to the target property of the k-level material index, the k-1-level index material set is retrieved to obtain the k-level index material set, where the initial value of k is 1, and the 0-level index material set is a flat material library; Determine whether the k-level index material set is greater than or equal to a material quantity threshold; When the k-level index material set is greater than or equal to the material quantity threshold, and k<N, k is increased by one, and the k-level index material set is returned to the material index module to execute a loop; when the k-level index material set is less than the material quantity threshold, or / and k=N, the k-level index material set is sent to the user terminal.

[0062] Furthermore, the embodiment of the present application also includes: Obtain a trait grading table; According to the trait grading table, traverse the material index target trait to obtain the target trait level identifier; According to the target trait level identifier, the material index target traits are grouped to obtain a plurality of groups of material index target traits, wherein the plurality of groups of material index target traits have a plurality of target trait level identifiers; According to the target property level identifiers, the groups of material index target properties are sorted from small to large levels to obtain the first-level material index target property, the second-level material index target property, and finally the N-level material index target property.

[0063] Furthermore, the embodiment of the present application also includes: When the plane material library is updated, the first property trigger frequency of the plane material library, the second property trigger frequency of the plane material library, and so on until the Mth property trigger frequency of the plane material library are obtained; Based on the trait trigger frequency deviation threshold, the material traits are clustered based on the first trait trigger frequency of the plane material library, the second trait trigger frequency of the plane material library, and up to the Mth trait trigger frequency of the plane material library to obtain a first group of trait sets, a second group of trait sets, and up to the Qth group of trait sets; Traversing the first set of trait sets, the second set of trait sets, and up to the Qth set of trait sets to perform trigger frequency mean calculation, and obtaining a first trigger frequency mean, a second trigger frequency mean, and up to the Qth trigger frequency mean; According to the first trigger frequency mean, the second trigger frequency mean, up to the Qth trigger frequency mean, the first group of trait sets, the second group of trait sets, up to the Qth group of trait sets are sorted from small to large to obtain a first level identification trait set, a second level identification trait set, up to the Lth level identification trait set; The trait grading table is constructed based on the first level identification trait set, the second level identification trait set, and up to the Lth level identification trait set.

[0064] Furthermore, the embodiment of the present application also includes: Obtaining a first attribute characteristic value set; Based on the trait characteristic value deviation threshold, cluster analysis is performed on the first trait attribute characteristic value set to obtain a trait attribute characteristic value clustering result; Any one of the clustering results of the trait attribute characteristic values ​​belongs to the first trait of the plane material library, the second trait of the plane material library, and up to the Mth trait of the plane material library, and the number of characteristic values ​​of any one of the clustering results of the trait attribute characteristic values ​​belongs to the first trait triggering frequency of the plane material library, the second trait triggering frequency of the plane material library, and up to the Mth trait triggering frequency of the plane material library; After each property attribute characteristic value set is analyzed, the first property trigger frequency of the plane material library, the second property trigger frequency of the plane material library, and so on until the Mth property trigger frequency of the plane material library are output.

[0065] Furthermore, the embodiment of the present application also includes: According to the k-1 level index material set, obtaining a k-1 level first index material; Extracting the k-1 level first index material having a k level material property feature value set; Receive k-level material index target trait deviation threshold; Counting the proportion of the number of trait attributes of the k-level material index target trait and the k-level material trait feature value set that is less than or equal to the k-level material index target trait deviation threshold, and setting it as k-level proximity; When the k-level proximity is greater than or equal to a proximity threshold, the k-1-level first index material is added to the k-level index material set.

[0066] Furthermore, the proximity threshold is equal to 1.

[0067] Embodiment three: See also Figure 3 , Figure 3 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, an electronic device 200 provided by an embodiment of the present invention includes a memory 210, a processor 220, and a first computer program 211 stored in the memory 210 and executable on the processor 220. When the processor 220 executes the first computer program 211, a custom retrieval graphic design material classification management system is implemented.

[0068] Embodiment 4: See also Figure 4 , Figure 4 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 4 As shown, this embodiment provides a computer-readable storage medium 300 on which a second computer program 311 is stored. When the second computer program 311 is executed by a processor, a graphic design material classification management system with customized retrieval is implemented.

[0069] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0070] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0071] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0072] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0073] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0074] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0075] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technology, the present invention is also intended to include these changes and variations.

Claims

1. A graphic design material classification management system with customized retrieval, characterized in that: include: A user end, the user end is used to receive a material index request, wherein the material index request includes a material index target property; The server includes: A material property grading module, the material property grading module is used to grade the material index target property to obtain a first-level material index target property, a second-level material index target property, and finally an N-level material index target property; A material indexing module, wherein the material indexing module is used to retrieve a k-1 level index material set according to a k level material index target property to obtain a k level index material set, where the initial value of k is 1, and the 0 level index material set is a plane material library; A loop determination module, the loop determination module is used to determine whether the k-level index material set is greater than or equal to a material quantity threshold; A loop decision module, wherein when the k-level index material set is greater than or equal to the material quantity threshold and k<N, k is increased by one, and the k-level index material set is returned to the material index module to execute a loop; when the k-level index material set is less than the material quantity threshold, or / and k=N, the k-level index material set is sent to the user terminal.

2. The system according to claim 1, characterized in that The material property classification module is used to classify the material index target property to obtain the first-level material index target property, the second-level material index target property, and finally the N-level material index target property, including: Obtain a trait grading table; According to the trait grading table, traverse the material index target trait to obtain the target trait level identifier; According to the target trait level identifier, the material index target traits are grouped to obtain a plurality of groups of material index target traits, wherein the plurality of groups of material index target traits have a plurality of target trait level identifiers; According to the target property level identifiers, the groups of material index target properties are sorted from small to large levels to obtain the first-level material index target property, the second-level material index target property, and finally the N-level material index target property.

3. The system according to claim 2, characterized in that The construction process of the trait grading table is as follows: When the plane material library is updated, the first property trigger frequency of the plane material library, the second property trigger frequency of the plane material library, and so on until the Mth property trigger frequency of the plane material library are obtained; Based on the trait trigger frequency deviation threshold, the material traits are clustered based on the first trait trigger frequency of the plane material library, the second trait trigger frequency of the plane material library, and up to the Mth trait trigger frequency of the plane material library to obtain a first group of trait sets, a second group of trait sets, and up to the Qth group of trait sets; Traversing the first set of trait sets, the second set of trait sets, and up to the Qth set of trait sets to perform trigger frequency mean calculation, and obtaining a first trigger frequency mean, a second trigger frequency mean, and up to the Qth trigger frequency mean; According to the first trigger frequency mean, the second trigger frequency mean, up to the Qth trigger frequency mean, the first group of trait sets, the second group of trait sets, up to the Qth group of trait sets are sorted from small to large to obtain a first level identification trait set, a second level identification trait set, up to the Lth level identification trait set; The trait grading table is constructed based on the first level identification trait set, the second level identification trait set, and up to the Lth level identification trait set.

4. The system according to claim 3, characterized in that When the plane material library is updated, obtaining the first property trigger frequency of the plane material library, the second property trigger frequency of the plane material library, and up to the Mth property trigger frequency of the plane material library includes: Obtaining a first attribute characteristic value set; Based on the trait characteristic value deviation threshold, cluster analysis is performed on the first trait attribute characteristic value set to obtain a trait attribute characteristic value clustering result; Any one of the clustering results of the trait attribute characteristic values ​​belongs to the first trait of the plane material library, the second trait of the plane material library, and up to the Mth trait of the plane material library, and the number of characteristic values ​​of any one of the clustering results of the trait attribute characteristic values ​​belongs to the first trait triggering frequency of the plane material library, the second trait triggering frequency of the plane material library, and up to the Mth trait triggering frequency of the plane material library; After each property attribute characteristic value set is analyzed, the first property trigger frequency of the plane material library, the second property trigger frequency of the plane material library, and so on until the Mth property trigger frequency of the plane material library are output.

5. The system according to claim 1, wherein: The material indexing module is used to retrieve a k-1 level index material set according to a k level material index target property to obtain a k level index material set, including: According to the k-1 level index material set, obtaining a k-1 level first index material; Extracting the k-1 level first index material having a k level material property feature value set; Receive k-level material index target trait deviation threshold; Counting the proportion of the number of trait attributes of the k-level material index target trait and the k-level material trait feature value set that is less than or equal to the k-level material index target trait deviation threshold, and setting it as k-level proximity; When the k-level proximity is greater than or equal to a proximity threshold, the k-1-level first index material is added to the k-level index material set.

6. The system according to claim 5, characterized in that The proximity threshold is equal to 1.

7. A method for classifying and managing graphic design materials with customized retrieval, characterized in that: include: receiving a material index request, wherein the material index request includes a material index target trait; The material index target traits are graded to obtain a first-level material index target trait, a second-level material index target trait, and finally an N-level material index target trait; According to the target property of the k-level material index, the k-1-level index material set is retrieved to obtain the k-level index material set, where the initial value of k is 1, and the 0-level index material set is a flat material library; Determine whether the k-level index material set is greater than or equal to a material quantity threshold; When the k-level index material set is greater than or equal to the material quantity threshold, and k<N, k is increased by one, and the k-level index material set is returned to the material index module to execute a loop; when the k-level index material set is less than the material quantity threshold, or / and k=N, the k-level index material set is sent to the user end.

8. The method according to claim 7, characterized in that The material index target trait is graded to obtain a first-level material index target trait, a second-level material index target trait, and finally an N-level material index target trait, including: Obtain a trait grading table; According to the trait grading table, traverse the material index target trait to obtain the target trait level identifier; According to the target trait level identifier, the material index target traits are grouped to obtain a plurality of groups of material index target traits, wherein the plurality of groups of material index target traits have a plurality of target trait level identifiers; According to the target property level identifiers, the groups of material index target properties are sorted from small to large levels to obtain the first-level material index target property, the second-level material index target property, and finally the N-level material index target property.

9. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing a custom search graphic design material classification management system as described in any one of claims 1-6.

10. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores a computer software program, which, when executed by a processor, implements a graphic design material classification management system with customized retrieval as described in any one of claims 1 to 6.

Images