Package design method and device based on target GNN model, equipment and medium

Abstract

The invention provides a package design method and device based on a target GNN model, equipment and a medium, is used for the technical field of package design, and can solve the problem that existing package design is poor in efficiency and quality. Comprising the steps of performing feature recognition on a to-be-designed image, and determining a target design vector according to each image vector and a corresponding target text vector; calculating a material similarity value and a process similarity value according to the target GNN model, and determining a target design scheme according to the material similarity value and the process similarity value; according to the user score value and the predicted score value corresponding to each target design scheme, calculating a score loss value, and according to a preset material, a preset process, the plurality of historical design schemes and the target design schemes, calculating a divergence value; performing weighted calculation on the score loss value and the divergence value to obtain a target loss value, and adjusting parameters of the target GNN model according to the target loss value and a preset loss threshold; therefore, the design quality and efficiency are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of packaging design, and particularly relates to a packaging design method and device based on a target GNN model, equipment and a medium. BACKGROUND

[0002] Packaging printing design is a deep combination of product packaging visual creativity, functional requirements and production landing, so as to create a packaging solution with protection function, brand communication and user experience through the integration of graphics, color, material and technology, so as to strengthen the advantages of products and improve competitiveness.

[0003] The existing packaging design method generally presets a packaging box structure, and the process recommendation depends on a rule base without fusing multi-modal image analysis. Or it matches materials through user labels, which is easy to ignore design details.

[0004] However, the existing packaging design method has the problems of low design efficiency and quality. SUMMARY

[0005] The present application provides a packaging design method and device based on a target GNN model, and a medium, to solve the problem of low design efficiency and quality of the existing packaging design method.

[0006] In a first aspect, the present application provides a packaging design method based on a target GNN model, the method comprising: performing feature recognition on a to-be-designed image to obtain each image feature and an image vector corresponding to the image feature, and determining a target design vector according to each image vector and a target text vector corresponding thereto; calculating a material similarity value of the target design vector and each material node vector and a process similarity value of the target design vector and each process node vector according to the target GNN model, and determining a plurality of target design schemes according to the material similarity value and the process similarity value; the target design scheme comprises a target material and a target process; calculating a score loss value according to a user score value and a predicted score value corresponding to each target design scheme, and calculating a divergence value according to a preset material, a preset process, a plurality of historical design schemes and the target design scheme; the predicted score value is a numerical value determined according to the target GNN model; weighting the score loss value and the divergence value according to a preset weight to obtain a target loss value, and adjusting parameters of the target GNN model according to the target loss value and a preset loss threshold to obtain an adjusted target GNN model, so that the target loss value calculated according to the adjusted target GNN model is less than the preset loss threshold.

[0007] In some embodiments of the present application, the target design vector is determined according to the respective image vector and the corresponding target text vector, comprising: performing semantic division on the text to be designed to obtain a plurality of subtexts, and determining the vector corresponding to each subtext to obtain a plurality of text vectors; determining the target text vector corresponding to each image vector in the text vector based on the contrast learning mechanism of the CLIP model, and associating the image vector and the target text vector to obtain a target vector; performing vector fusion on all target vectors and text vectors according to the attention mechanism of the CLIP model to obtain a target design vector.

[0008] In some embodiments of the present application, before calculating the material similarity value of the target design vector with respect to each material node vector and the process similarity value with respect to each process node vector according to the target GNN model, the method further comprises: determining the corresponding sample design vector according to the CLIP model, the sample image and the corresponding sample text, and determining the positive sample scheme and the negative sample scheme corresponding to the sample design vector according to the preset design database; the positive sample scheme is a sample scheme associated with the sample design vector, and the negative sample scheme is a sample scheme not associated with the sample design vector; determining the positive sample vector corresponding to the positive sample scheme and the negative sample vector corresponding to the negative sample scheme, and calculating the positive similarity value of the sample design vector with respect to the positive sample vector and the negative similarity value with respect to the negative sample vector according to the initial GNN model; determining the corresponding loss value according to the difference between the negative similarity value and the positive similarity value, and adjusting the model parameters of the initial GNN model according to the loss value and the preset threshold to obtain the target GNN model.

[0009] In some embodiments of the present application, adjusting the model parameters of the initial GNN model according to the loss value and the preset threshold to obtain the target GNN model, comprising: comparing the loss value and the preset threshold to obtain the corresponding comparison result; if the comparison result is that the loss value is less than the preset threshold, then adjusting the model parameters of the initial GNN model according to the loss value to obtain the target GNN model, so that the loss value calculated according to the target GNN model is not less than the preset threshold; if the comparison result is that the loss value is not less than the preset threshold, then determining the initial GNN model as the target GNN model.

[0010] In some embodiments of the present application, the plurality of target design schemes are determined according to the material similarity value and the process similarity value, comprising: According to a preset corresponding rule, at least one process node corresponding to each material node is determined, and each material node and the corresponding process node are determined as a target design node; the material node is a node corresponding to a material node vector, and the process node is a node corresponding to a process node vector; The target material similarity value corresponding to each target design node and the target process similarity value are added to obtain a plurality of design similarity values, and the design similarity values are sorted from large to small according to the values to obtain a design similarity value sequence; the target material similarity value is a material similarity value corresponding to a material node in the target design node, and the target process similarity value is a process similarity value corresponding to a process node in the target design node; A preset number of design similarity values in the design similarity value sequence are sequentially determined to obtain a plurality of target design similarity values, and according to the target material node and the target process node corresponding to each target design similarity value, the corresponding target material and target process are determined. According to each target material and target process, a plurality of corresponding target design schemes are determined.

[0011] In some embodiments of the present application, according to the user score value and the predicted score value corresponding to each target design scheme, a corresponding score loss value is calculated, including: A plurality of target design schemes are determined and output; According to the output feedback result corresponding to the target design scheme, a user score value corresponding to each target design scheme is determined, and according to the target GNN model, a predicted score value corresponding to each target design scheme is determined; According to the user score value and the corresponding predicted score value, an error value corresponding to each target design scheme is calculated, and an average value corresponding to all error values is calculated to obtain a score loss value.

[0012] In some embodiments of the present application, according to the preset material, the preset process, the plurality of historical design schemes and the target design scheme, a divergence value is calculated, including: According to the preset design database, the preset material, the preset process and the plurality of historical design schemes are determined, and according to the preset material and the preset process, a target material scheme and a target process scheme in the historical design scheme and the target design scheme are respectively determined; the target material scheme is a scheme containing the preset material, and the target process scheme is a scheme containing the preset process; According to the target material quantity corresponding to the target design scheme and the historical design scheme respectively, a material divergence value is calculated, and according to the target process quantity corresponding to the target design scheme and the historical design scheme respectively, a process divergence value is calculated; the target material quantity is a quantity corresponding to the target material scheme, and the target process quantity is a quantity corresponding to the target process scheme; According to the preset weight values corresponding to the target material quantity and the target process quantity respectively, the target material quantity and the target process quantity are weighted and summed to obtain a divergence value.

[0013] In a second aspect, the application provides a packaging design device based on a target GNN model, the device comprising: An identification module is configured to perform feature recognition on the to-be-designed images to obtain image features and image vectors corresponding to the image features, and determine a target design vector according to the image vectors and target text vectors corresponding to the image vectors; A determination module is configured to calculate material similarity values between the target design vector and material node vectors and process similarity values between the target design vector and process node vectors according to the target GNN model, and determine a plurality of target design schemes according to the material similarity values and the process similarity values; the target design schemes comprise target materials and target processes; A calculation module is configured to calculate a score loss value according to user score values and predicted score values corresponding to each target design scheme, and calculate a divergence value according to preset materials, preset processes, a plurality of historical design schemes and the target design schemes; the predicted score values are numerical values determined according to the target GNN model; An adjustment module is configured to perform weighted calculation on the score loss value and the divergence value according to a preset weight to obtain a target loss value, and adjust parameters of the target GNN model according to the target loss value and a preset loss threshold to obtain an adjusted target GNN model, so that the target loss value calculated according to the adjusted target GNN model is less than the preset loss threshold.

[0014] In a third aspect, the application provides a computer device, comprising a processor and a memory in communication connection with the processor; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory to implement the method of the application.

[0015] In a fourth aspect, the application provides a computer readable storage medium, the computer readable storage medium storing program codes, the program codes being executed by a processor to implement the method of the application.

[0016] The application provides a packaging design method and device based on a target GNN model, which comprises the following steps: performing feature recognition on a to-be-designed image to obtain image features and image vectors corresponding to the image features; determining a target design vector according to the image vectors and target text vectors corresponding to the image vectors; calculating material similarity values of the target design vector and material node vectors and process similarity values of the target design vector and process node vectors according to the target GNN model, and determining a plurality of target design schemes according to the material similarity values and the process similarity values; the target design scheme comprises a target material and a target process; calculating a score loss value according to user score values and predicted score values corresponding to the target design schemes, and calculating a divergence value according to preset materials, preset processes, a plurality of historical design schemes and the target design schemes; the predicted score value is a value determined according to the target GNN model; performing weighted calculation on the score loss value and the divergence value according to a preset weight to obtain a target loss value, and adjusting parameters of the target GNN model according to the target loss value and a preset loss threshold to obtain an adjusted target GNN model, so that the target loss value calculated according to the adjusted target GNN model is less than the preset loss threshold.

[0017] In this way, by performing feature recognition on the to-be-designed image and generating image vectors, the target design vector is determined in combination with the target text vector, so that the accurate recognition of image features is realized according to the image, and the image features and corresponding text features are further fused for feature, so that the matching between the materials and the processes is realized, and the actual design is prevented from being disconnected from the user demand; by the GNN model, the similarity values of the target design vector and the material and process node vectors are calculated, and the scheme containing the target material and the process is generated, so that the correlation reasoning ability of the graph neural network is utilized to efficiently match the optimal material and process combination, and multiple scheme selection is provided to solve the problems of low matching efficiency of the materials and the processes and single generated scheme; by calculating the score loss value, the fitting degree of the quantitative model to the user preference can be realized, and the divergence value can be calculated to constrain the consistency of the generated scheme with the long-term style and basic requirements of the enterprise, so that the loss value is calculated according to the combination of the two, which can ensure that the design meets the real-time demand of the user and avoids deviating from the established strategy of the enterprise; by the preset weight, the target loss value is obtained by weighting the score loss value and the divergence value, and the GNN model parameters are adjusted accordingly, so that the loss value is lower than the threshold, and the continuous iterative optimization of the model can be realized, so that the accuracy of the subsequent design scheme in meeting the user demand, matching the enterprise style and adapting to the materials and processes is continuously improved, and the efficiency, accuracy and design quality of the packaging design are improved. BRIEF DESCRIPTION OF DRAWINGS

[0018] The accompanying drawings, which are incorporated into and form a part of the specification, illustrate one embodiment consistent with the present application and, together with the specification, serve to explain the principles of the application.

[0019] Figure 1 A flowchart of a packaging design method based on a target GNN model provided by an embodiment of the present application is shown in FIG. 1. Figure 2 A flowchart of another packaging design method based on a target GNN model provided by an embodiment of the present application is shown in FIG. 2. Figure 3 A structural diagram of a packaging design device based on a target GNN model provided by an embodiment of the present application is shown in FIG. 3. Figure 4 A structural diagram of a device for performing a packaging design method based on a target GNN model according to an embodiment of the present application is shown in FIG. 4. DETAILED DESCRIPTION

[0020] The exemplary embodiments will be described in detail herein with reference to the attached drawings. The following description is with reference to the drawings, in which like numerals indicate like elements, unless otherwise specified. The embodiments described in the following exemplary embodiments are not meant to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the present application as detailed in the appended claims.

[0021] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the present application will be described below with reference to the drawings.

[0022] Figure 1 A flowchart of a packaging design method based on a target GNN model provided by an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the packaging design method based on a target GNN model can include the following steps: Figure 1 S110, feature recognition is performed on the image to be designed to obtain image features and image vectors corresponding to the image features, and a target design vector is determined according to the image vectors and their corresponding target text vectors.

[0023] The image to be designed refers to the original image material that needs to be designed for packaging and printing, which is usually a two-dimensional design drawing, such as a product packaging plan view, a design draft containing a bottom line or a logo or a decorative element, etc., thereby providing visual basic information for packaging design, so as to meet the real-time design needs of users according to the image to be designed.

[0024] ​Image features refer to visual areas with specific semantics identified and segmented from the image to be designed. They can be detected by a YOLO detection model, thereby converting the original image into quantifiable visual elements to realize the conversion of image information into computable features. For example, the base color type can be identified as a specific feature in the image, such as a red matte gradient or a gold frosted image.

[0025] Target text vector refers to the high-dimensional vector converted from the user's text requirements for packaging design, thereby realizing the vectorization expression of the text for subsequent calculation based on the vector.

[0026] Target design vector refers to the comprehensive vector obtained by fusing the image vector corresponding to each image feature and the target text vector corresponding to the image feature, which contains both visual information of the image and semantic requirements of the text. It can be directly input into the subsequent GNN model for calculating the similarity with the material node vector and the process node vector, thereby ensuring the consistency of the final design scheme in terms of vision and function.

[0027] Based on this, by determining the image vector corresponding to the image to be designed and the text vector corresponding to the design text, the combination of image features and text features is realized through the fusion of image vectors and text vectors, so as to ensure that the target scheme generated finally can take into account both image and text requirements, thereby meeting the design requirements of users and improving the quality of design.

[0028] S120, according to the target GNN model, calculating the material similarity value of the target design vector and each material node vector, and the process similarity value of the target design vector and each process node vector, and determining a plurality of target design schemes according to the material similarity value and the process similarity value; the target design scheme includes target material and target process.

[0029] Among them, the target GNN model refers to a GNN model pre-trained in a specific way, so as to realize the mapping of the input target design vector to an embedding space and the calculation of the similarity between the vector and all node vectors in the knowledge base; the GNN model refers to a graph neural network (Graph Neural Network) model, which is an algorithm framework for processing graph structure data based on deep learning, Material node vector refers to a high-dimensional vector representation of various design-related packaging materials, which is generated by encoding the attributes of the materials, such as material, thickness, color, scratch resistance, cost, etc., thereby quantifying the features of the materials and converting the attributes of the physical materials into a vector form that can be calculated by a computer.

[0030] The material similarity value refers to the similarity score of the target design vector and a certain material node vector, which is usually a value between 0 and 1, and can be calculated by cosine similarity, Euclidean distance, etc. The higher the value, the higher the matching degree of the material and the design requirement.

[0031] The process node vector refers to the high-dimensional vector representation of various printing processes in the process library, which is generated by encoding the attributes of the process, such as implementation effect, applicable material, cost, precision, etc., so as to quantify the characteristics of the process, so as to convert the abstract process effect into a calculable vector.

[0032] The process similarity value refers to the similarity score of the target design vector and a certain process node vector, which is usually a value between 0 and 1, and can be calculated by cosine similarity, Euclidean distance, etc. The higher the value, the higher the matching degree of the process and the design requirement.

[0033] The target design scheme refers to the combination scheme of materials and processes selected according to the material similarity value and the process similarity value, such as 275g photo paper, silk screen crystal oil and knock-out, which includes specific material selection, process type and parameters, so as to convert the abstract design requirement into a producible production scheme.

[0034] The target material refers to the specific packaging material contained in the target design scheme, and the target process refers to the specific packaging process contained in the target design scheme.

[0035] Therefore, by using the target GNN model pre-trained in a specific way, the similarity values between the target design vector and a plurality of predetermined material node vectors and process node vectors are calculated, so as to determine the target material and target process that best match the target design vector according to the numerical value of the similarity value.

[0036] In S130, the score loss value is calculated according to the user score value and the predicted score value corresponding to each target design scheme, and the divergence value is calculated according to the preset material, the preset process, the plurality of historical design schemes and the target design scheme. The predicted score value is a value determined according to the target GNN model.

[0037] The user score value refers to the actual evaluation score of the user for the generated target design scheme, reflecting the user's satisfaction with the scheme, providing an intuitive basis for subsequent adjustment of the model preference, avoiding the generation of the model from deviating from the user's demand, and thus improving the design quality.

[0038] The predicted score value refers to the score value predicted in advance by the target GNN model according to the association relationship between the design scheme and the user preference learned by the model, which is used to measure the understanding accuracy of the model to the user preference, so as to realize the prediction of the user's welcome degree of the scheme by the model.

[0039] The score loss value refers to a loss index calculated by quantifying the difference between the predicted score value and the user score value, and a commonly used calculation method is mean square error.

[0040] The preset material refers to a packaging material that is pre-set in the technical solution and meets the basic needs or hard constraints of the enterprise, and is usually a conventional material provided by a long-term cooperative supplier of the enterprise; and the preset process refers to a packaging process that is pre-set in the technical solution and meets the basic needs or hard constraints of the enterprise.

[0041] The historical design scheme refers to a set of packaging design schemes that have been completed before or have good user feedback The divergence value refers to an index calculated by quantifying the difference between the distribution of materials and processes in the target design scheme and the distribution of materials and processes in the historical design scheme, so as to prevent the model from deviating from the long-term style to cater to a single score, that is, by calculating the schemes containing preset materials and preset processes in the historical design scheme and the target design scheme, the corresponding divergence value is calculated.

[0042] Based on this, after generating the target design scheme, the loss value is calculated by determining the different score values of the scheme between the user and the model, and the divergence value is calculated by calculating the respective appearance frequencies of materials and processes in the historical scheme and the target scheme this time, so as to further adjust the parameters of the target GNN model according to the loss value and the divergence value in the subsequent, thereby improving the quality of the generated scheme.

[0043] S140, according to the preset weight, the score loss value and the divergence value are weighted and calculated to obtain a target loss value, and according to the target loss value and a preset loss threshold, the parameters of the target GNN model are adjusted to obtain an adjusted target GNN model, so that the target loss value calculated according to the adjusted target GNN model is less than the preset loss threshold.

[0044] The preset weight is a pre-determined weight value corresponding to the score loss value and the divergence value, respectively, and is used to balance the influence degree of the score loss value and the divergence value in the total loss.

[0045] The target loss value refers to an overall comprehensive loss value obtained by multiplying the score loss value and its corresponding weight value, and multiplying the divergence value and its corresponding weight value, and then adding the two multiplication values, which is a comprehensive quantitative evaluation of the model prediction deviation and the style deviation degree, and the smaller the value is, the better the comprehensive performance of the model in meeting the user's demand and meeting the enterprise's constraints.

[0046] The preset loss threshold refers to a pre-set loss threshold value for judging whether the target GNN model reaches the optimization target, and when the target loss value is less than the threshold value, it means that the model's comprehensive performance meets the standard; if it is greater than the threshold value, the parameters need to be continuously adjusted.

[0047] Based on this, the score loss value can represent the different score values between the target design scheme and the user and the model, the divergence value can represent the respective distribution rates of the preset material and the preset process in the historical design scheme and the target design scheme, and the target design scheme itself is determined by the target GNN, so that the target loss value is obtained by weighting calculation on the score loss value and the divergence value, and the model parameters of the target GNN model are adjusted according to the target loss value, so that the adjusted GNN model can better meet the actual needs of the user.

[0048] On the basis of the above feasible implementation manner of S110, the present application further provides that the target design vector is determined according to each image vector and the corresponding target text vector, comprising: performing semantic division on the to-be-designed text to obtain a plurality of subtexts, and determining a vector corresponding to each subtext to obtain a plurality of text vectors; determining the target text vector corresponding to each image vector in the text vector based on the contrast learning mechanism of the CLIP model, and associating the image vector and the target text vector to obtain a target vector; performing vector fusion on all target vectors and text vectors according to the attention mechanism of the CLIP model to obtain a target design vector.

[0049] The to-be-designed text refers to the user's textual description requirement for packaging and printing design, including key information such as function, style, and scene.

[0050] The CLIP (Contrastive Language-Image Pretraining) model is a cross-modal pre-training model, which is trained by large-scale image and text pair data to realize the semantic association between image and text. The model can understand the corresponding relationship between the visual content of the image and the semantic description of the text, such as the matching between the red gift box text and the red gift box image.

[0051] The target vector refers to a comprehensive vector formed by associating a single image vector with its matching target text vector. For example, after associating the image vector of red undercoat with the target text vector of wedding theme, a fusion vector containing visual red and semantic wedding is formed.

[0052] Based on this, after determining the image feature corresponding to the image vector, the text can be divided according to the semantics of the to-be-designed text, and each subtext is expressed as a vector to obtain a corresponding text vector. Further, through the CLIP model, the cross-modal semantic association between text and image is realized through contrastive learning, so that both image and text can be processed and fused to obtain a target design vector.

[0053] Based on the feasible implementation of S120 described above, this application further provides a method that, before calculating the material similarity value between the target design vector and each material node vector, and the process similarity value with each process node vector according to the target GNN model, includes: Based on the CLIP model, sample images, and corresponding sample text, the corresponding sample design vector is determined. Then, based on the preset design database, the positive and negative sample schemes corresponding to the sample design vector are determined. The positive sample scheme is the sample scheme that is associated with the sample design vector, and the negative sample scheme is the sample scheme that is not associated with the sample design vector. Determine the positive sample vector corresponding to the positive sample scheme and the negative sample vector corresponding to the negative sample scheme, and calculate the positive similarity value between the sample design vector and the positive sample vector and the negative similarity value between the sample design vector and the negative sample vector based on the initial GNN model. Based on the difference between negative and positive similarity values, the corresponding loss value is determined. Then, based on the loss value and a preset threshold, the model parameters of the initial GNN model are adjusted to obtain the target GNN model.

[0054] Among them, sample images refer to historical packaging design images used to train the model; sample text refers to the text descriptions corresponding to the sample images; and sample design vectors refer to the composite vectors obtained by fusing the image vectors of the sample images and the text vectors of the sample texts through the CLIP model.

[0055] The pre-built design database refers to a pre-constructed database containing a large number of historical packaging design schemes. Each scheme records information such as materials, processes, and corresponding design requirements.

[0056] A positive sample solution refers to a historical design solution in a pre-defined design database that is strongly correlated with the sample design vector. For example, if the sample design vector is "wedding, red", the positive sample solution is "wedding gift box with red background and hot stamping", which is a valid solution that meets the sample design requirements.

[0057] Negative sample schemes refer to historical design schemes in the pre-set design database that are unrelated or weakly related to the sample design vector. For example, if the sample design vector is "wedding, red", the negative sample scheme is "blue background + ordinary offset printing industrial packaging scheme", which is an invalid scheme that does not meet the sample design requirements.

[0058] The preset threshold refers to the critical value set during model training to determine whether the difference between positive and negative similarity values ​​meets the standard.

[0059] Based on this, in order to realize the specific pre-training of the GNN model, the sample data, i.e., the sample design vector, can be determined, and the positive sample scheme associated with the sample design vector and the negative sample scheme not associated with the sample design vector are determined. The positive sample scheme represents the scheme corresponding to the current design requirement, and the negative sample scheme is the scheme not corresponding to the design requirement. By determining the corresponding positive sample vector and negative sample vector, and according to the initial GNN model, the positive similarity value of the sample design vector and the positive sample vector, and the negative similarity value of the sample design vector and the negative sample vector are calculated. Correspondingly, the positive similarity value can be understood as the similarity value that meets the user's demand, and the negative similarity value can be understood as the similarity value that does not meet the user's demand. In actual application, in order to improve the calculation accuracy of the GNN model and thus improve the quality of the packaging design, the result output by the GNN model is more similar to the user's demand. Therefore, by calculating the difference between the negative similarity value and the positive similarity value, the corresponding loss value is obtained, which can reflect the difference between the sample vector generated by the GNN model and the positive scheme and the negative scheme determined by the user. By adjusting the loss value to be higher than the preset threshold, the optimization of the GNN model is realized, and the target GNN model is obtained.

[0060] Based on the above feasible implementation manner of S120, the application further provides adjusting the model parameters of the initial GNN model according to the loss value and the preset threshold to obtain the target GNN model, comprising: comparing the loss value and the preset threshold to obtain a corresponding comparison result; if the comparison result is that the loss value is less than the preset threshold, adjusting the model parameters of the initial GNN model according to the loss value to obtain the target GNN model, so that the loss value calculated according to the target GNN model is not less than the preset threshold; if the comparison result is that the loss value is not less than the preset threshold, determining the initial GNN model as the target GNN model.

[0061] Based on this, after obtaining the loss value, since the loss value can represent the difference between the positive similarity value and the negative similarity value, i.e., the similarity difference between the schemes generated by the GNN model, by determining the preset threshold, the loss value and the preset threshold are compared, so as to determine the target GNN model that meets the actual demand according to different comparison results.

[0062] Based on the above feasible implementation manner of S120, the application further provides determining a plurality of target design schemes according to the material similarity value and the process similarity value, comprising: According to a preset correspondence rule, at least one process node corresponding to each material node is determined, and each material node and the corresponding process node are determined as a target design node; the material node is a node corresponding to a material node vector, and the process node is a node corresponding to a process node vector; The target material similarity value of each target design node and the target process similarity value are added to obtain a plurality of design similarity values, and the design similarity values are sorted from large to small according to the numerical values to obtain a design similarity value sequence; the target material similarity value is a material similarity value corresponding to a material node in the target design node, and the target process similarity value is a process similarity value corresponding to a process node in the target design node; A preset number of design similarity values in the design similarity value sequence are sequentially determined to obtain a plurality of target design similarity values, and according to the target material node and the target process node corresponding to each target design similarity value, the corresponding target material and target process are determined. According to each target material and target process, a plurality of corresponding target design schemes are determined.

[0063] Among them, the preset correspondence rule refers to a rule that is pre-set and restricts the association relationship between the material node and the process node, and is used to screen the combination of materials and processes that are technically feasible, adaptively effective, and between materials and processes, thereby excluding invalid combinations.

[0064] The target design node refers to a combination unit of the material node and the process node associated through the preset correspondence rule, and is the smallest representation unit representing the matching relationship between the material and the process; for example, the material node of 275g photo paper and the process node of silk printing crystal oil are combined to form a target design node.

[0065] The target material similarity value refers to the similarity value of the material node in the target design node and the target design vector; the target process similarity value refers to the similarity value of the process node in the target design node and the target design vector.

[0066] The preset number refers to a pre-set number of target design schemes that need to be finally output.

[0067] The target material node refers to the material node in the target design node corresponding to the target design similarity value; the target process node refers to the process node in the target design node corresponding to the target design similarity value.

[0068] The target material refers to the actual packaging material corresponding to the target material node; the target process refers to the actual packaging process corresponding to the target process node.

[0069] Based on this, in actual application, the materials and processes of packaging design need to meet the actual corresponding rules, that is, the selected materials and processes must be adaptive, so by determining the processes corresponding to each material first, the materials and processes that meet the corresponding relationship are screened out to obtain the target design node. Since the similarity value represents the similarity between the target design vector and each material node vector and each process node vector, and the material node vector and the process node are respectively corresponding to the material node and the process node, the design similarity value corresponding to the target design node can be calculated by adding the sum according to the material node and the process node contained in each target design node. The design similarity value represents the similarity between the material and the process and the actual design requirement of the user (i.e. the target design vector), so according to the numerical value from large to small, the material and process highly similar to the target design vector are determined to obtain multiple target design schemes.

[0070] Based on the above feasible implementation manner of S130, the application further provides calculating a score loss value according to the user score value and the predicted score value corresponding to each target design scheme, comprising: determining multiple target design schemes and outputting; determining the user score value corresponding to each target design scheme according to the output feedback result corresponding to the target design scheme, and determining the predicted score value corresponding to each target design scheme according to the target GNN model; calculating the error value corresponding to each target design scheme according to the user score value and the corresponding predicted score value, and calculating the average value corresponding to all error values to obtain the score loss value.

[0071] Among them, the output feedback result refers to the subjective evaluation given by the user based on the visual effect, functional adaptability, production feasibility, cost acceptability and other dimensions of the scheme after the output of the target design scheme. For example, it can include user score value, such as quantitative scoring of 0-10 points, and can also cover qualitative feedback such as text suggestions and scheme selection results.

[0072] Based on this, by outputting multiple target design schemes, the user score value corresponding to each scheme is determined according to the feedback operation of the user in order to determine the design tendency of each scheme of the user, and the predicted score value corresponding to each scheme is predicted according to the target GNN model in order to determine whether the prediction of the target GNN model meets the actual requirements according to the error value between the user score value and the predicted score value, and calculate the average value corresponding to all error values to obtain the score loss value. In actual application, the loss calculation can be performed by mean square error loss function (MSE). MSE is the most commonly used loss function in regression problems, which is used to measure the difference between the predicted value of the model and the true value, that is, the prediction accuracy of the model is evaluated by calculating the average value of the prediction error square.

[0073] On the basis of the above feasible implementation of S130, the present application further provides a divergence value according to the preset material, the preset process, a plurality of historical design schemes and a target design scheme, comprising: According to the preset design database, the preset material, the preset process and the plurality of historical design schemes are determined, and the target material scheme and the target process scheme in the historical design scheme and the target design scheme are determined according to the preset material and the preset process; the target material scheme is a scheme containing the preset material, and the target process scheme is a scheme containing the preset process; According to the target material quantity corresponding to the target design scheme and the historical design scheme respectively, the material divergence value is calculated, and according to the target process quantity corresponding to the target design scheme and the historical design scheme respectively, the process divergence value is calculated; the target material quantity is the quantity corresponding to the target material scheme, and the target process quantity is the quantity corresponding to the target process scheme; According to the preset weight value corresponding to the target material quantity and the target process quantity respectively, the target material quantity and the target process quantity are weighted and summed to obtain the divergence value.

[0074] The material divergence value refers to the divergence index calculated by quantifying the number of schemes containing the preset material in the target design scheme, that is, the target material quantity, and the distribution difference of the number of schemes containing the preset material in the historical design scheme, that is, the historical target material quantity, which is commonly calculated by KL divergence, cosine distance and the like; for example, there are 100 historical design schemes, of which 80 contain 275g of lithography paper as the preset material, so the historical target material quantity accounts for 80%; there are 5 target design schemes, of which 2 contain the preset material, so the target material quantity accounts for 40%; the material divergence value is used to measure the distribution deviation degree of 40% and 80%, the smaller the value, the closer the distribution, and then the target scheme is more in line with the historical material preference.

[0075] The process divergence value is the divergence index calculated by quantifying the number of schemes containing the preset process in the target design scheme, that is, the target process quantity, and the distribution difference of the number of schemes containing the preset process in the historical design scheme, that is, the historical target process quantity.

[0076] Therefore, in practical applications, to prevent the model from deviating from its long-term style to cater to a single score, such as a company specializing in wedding products whose packaging should maintain its preferred gold foil tradition, a preset database can be established to determine preset materials and processes. Based on multiple randomly selected historical design schemes and multiple target design schemes that include preset materials or processes, material divergence values ​​and process divergence values ​​can be calculated. Since the target design schemes are generated based on the target GNN model, the model parameters can be adjusted by adjusting the divergence value obtained after weighted summation of the material divergence value and process divergence value.

[0077] Please refer to Figure 2 , Figure 2 A flowchart illustrating another packaging design method based on a target GNN model provided in this application embodiment; as shown Figure 2 As shown, by inputting a two-dimensional design drawing and requirement text, the region segmentation unit under the multimodal analysis module uses YOLO to segment elements such as background, logo, and decoration. The image-text alignment unit generates a semantic vector Vd using the CLIP model, and the knowledge matching unit calculates similarity using GNN. Then, the solution generation module enters the solution generation module. The material and process combination unit outputs the Top-K solutions, the cost calculation unit calls local knowledge values, and then uses the Blender physics engine to perform 3D effect rendering, thereby outputting a three-dimensional rendering and cost list. User rating feedback is collected, and the GNN weights are updated based on the feedback through the closed-loop learning unit.

[0078] In some embodiments of this application, feature recognition is performed on the image to be designed to obtain various image features and corresponding image vectors. Based on each image vector and its corresponding target text vector, a target design vector is determined. According to the target GNN model, the material similarity value between the target design vector and each material node vector, and the process similarity value between the target design vector and each process node vector, are calculated. Based on the material similarity value and the process similarity value, multiple target design schemes are determined. Each target design scheme includes a target material and a target process. Based on the user rating value and predicted rating value corresponding to each target design scheme, a rating loss value is calculated. Based on preset materials, preset processes, multiple historical design schemes, and the target design scheme, a divergence value is calculated. The predicted rating value is a value determined according to the target GNN model. Based on preset weights, the rating loss value and the divergence value are weighted to obtain a target loss value. Based on the target loss value and a preset loss threshold, the parameters of the target GNN model are adjusted to obtain an adjusted target GNN model, so that the target loss value calculated based on the adjusted target GNN model is less than the preset loss threshold.

[0079] Thus, by feature recognition on the image to be designed and generation of an image vector, the target design vector is determined in combination with the target text vector, so as to realize accurate recognition of image features according to the image, and further feature fusion of image features and corresponding text features, so as to match the subsequent materials and processes, while avoiding the disconnection between the actual design and the user demand; and by the GNN model, the similarity values of the target design vector and the material and process node vectors are calculated, and a scheme containing the target material and process is generated, which can not only utilize the correlation reasoning capability of the graph neural network to efficiently match the optimal material and process combination, but also provide multiple scheme options to solve the problems of low matching efficiency of materials and processes and single generated scheme; by calculating the scoring loss value, the fitting degree of the quantitative model to the user preference can be realized, and the calculation of the divergence value can constrain the consistency of the generated scheme with the long-term style and basic requirements of the enterprise, so as to calculate the loss value according to the combination of the two, which can not only ensure that the design meets the real-time demand of the user, but also avoids deviation from the established strategy of the enterprise; by presetting the weight, the target loss value is obtained by weighting the scoring loss value and the divergence value, and the GNN model parameters are adjusted accordingly, so that the loss value is lower than the threshold value, and the continuous iterative optimization of the model can be realized, so as to continuously improve the accuracy of the subsequent design scheme in meeting the user demand, matching the enterprise style, and adapting to the materials and processes, and improve the efficiency, accuracy and design quality of the packaging design.

[0080] Figure 3 A structure schematic diagram of a packaging design device 300 based on a target GNN model is provided for the embodiments of the present application. As shown in the figure, Figure 3 The packaging design device 300 based on the target GNN model includes an identification module 310, a determination module 320, a calculation module 330, and an adjustment module 340; wherein: The identification module 310 is configured to perform feature recognition on the image to be designed to obtain each image feature and an image vector corresponding to the image feature, and determine a target design vector according to each image vector and a target text vector corresponding thereto; The determination module 320 is configured to calculate material similarity values of the target design vector and each material node vector, and process similarity values of the target design vector and each process node vector according to the target GNN model, and determine a plurality of target design schemes according to the material similarity values and the process similarity values; the target design scheme includes a target material and a target process; The calculation module 330 is configured to calculate a scoring loss value according to a user score value and a predicted score value corresponding to each target design scheme, and calculate a divergence value according to a preset material, a preset process, a plurality of historical design schemes, and the target design scheme; the predicted score value is a value determined according to the target GNN model; The adjusting module 340 is configured to: perform weighted calculation on the score loss value and the divergence value according to a preset weight, to obtain a target loss value; and perform adjustment on parameters of the target GNN model according to the target loss value and a preset loss threshold, to obtain an adjusted target GNN model, so that a target loss value calculated according to the adjusted target GNN model is less than the preset loss threshold.

[0081] In the embodiments of the present application, the identifying module 310 can be specifically configured to: perform semantic division on the to-be-designed text to obtain a plurality of subtexts, and determine vectors corresponding to the subtexts to obtain a plurality of text vectors; determine, based on a contrast learning mechanism of a CLIP model, target text vectors corresponding to the image vectors in the text vectors, and associate the image vectors and the target text vectors to obtain target vectors; perform vector fusion on all the target vectors and the text vectors according to an attention mechanism of the CLIP model to obtain a target design vector.

[0082] In the embodiments of the present application, the determining module 320 can be specifically configured to: determine a sample design vector corresponding to the CLIP model, the sample image and the corresponding sample text, and determine, according to a preset design database, a positive sample scheme and a negative sample scheme corresponding to the sample design vector; the positive sample scheme is a sample scheme having an association with the sample design vector, and the negative sample scheme is a sample scheme having no association with the sample design vector; determine a positive sample vector corresponding to the positive sample scheme and a negative sample vector corresponding to the negative sample scheme, and calculate, according to the initial GNN model, a positive similarity value of the sample design vector with respect to the positive sample vector and a negative similarity value of the sample design vector with respect to the negative sample vector; determine a loss value corresponding to a difference between the negative similarity value and the positive similarity value, and adjust model parameters of the initial GNN model according to the loss value and a preset threshold to obtain a target GNN model.

[0083] In the embodiments of the present application, the determining module 320 can be specifically configured to: compare the loss value with the preset threshold to obtain a corresponding comparison result; if the comparison result is that the loss value is less than the preset threshold, adjust the model parameters of the initial GNN model according to the loss value to obtain the target GNN model, so that a loss value calculated according to the target GNN model is not less than the preset threshold; if the comparison result is that the loss value is not less than the preset threshold, determine that the initial GNN model is the target GNN model.

[0084] In the embodiments of the present application, the determining module 320 can be specifically configured to: According to a preset correspondence rule, at least one process node corresponding to each material node is determined, and each material node and the corresponding process node are determined as a target design node; the material node is a node corresponding to the material node vector, and the process node is a node corresponding to the process node vector; The target material similarity value and the target process similarity value corresponding to each target design node are added to obtain a plurality of design similarity values, and the design similarity values are sorted from large to small according to the numerical values to obtain a design similarity value sequence; the target material similarity value is a material similarity value corresponding to the material node in the target design node, and the target process similarity value is a process similarity value corresponding to the process node in the target design node; A preset number of design similarity values in the design similarity value sequence are sequentially determined to obtain a plurality of target design similarity values, and according to the target material node and the target process node corresponding to each target design similarity value, the corresponding target material and target process are determined; According to each target material and target process, a plurality of corresponding target design schemes are determined.

[0085] In the embodiments of the present application, the calculation module 330 can be specifically used for: determining a plurality of target design schemes and outputting; determining a user score value corresponding to each target design scheme according to an output feedback result corresponding to the target design scheme, and determining a predicted score value corresponding to each target design scheme according to the target GNN model; According to the user score value and the corresponding predicted score value, the error value corresponding to each target design scheme is calculated, and the average value corresponding to all error values is calculated to obtain a score loss value.

[0086] In the embodiments of the present application, the calculation module 330 can be specifically used for: According to a preset design database, a preset material, a preset process and a plurality of historical design schemes are determined, and target material schemes and target process schemes in the historical design schemes and the target design schemes are respectively determined according to the preset material and the preset process; the target material scheme is a scheme containing the preset material, and the target process scheme is a scheme containing the preset process; According to the target material quantity corresponding to the target design scheme and the historical design scheme respectively, a material divergence value is calculated, and according to the target process quantity corresponding to the target design scheme and the historical design scheme respectively, a process divergence value is calculated; the target material quantity is a quantity corresponding to the target material scheme, and the target process quantity is a quantity corresponding to the target process scheme; According to the preset weight value corresponding to the target material quantity and the target process quantity respectively, the target material quantity and the target process quantity are weighted and summed to obtain a divergence value.

[0087] Figure 4 A structural schematic diagram of an apparatus for performing a target GNN model-based packaging design method according to an embodiment of the present application is provided for the embodiment of the present application. As shown in the figure, the apparatus 400 includes: Figure 4 The apparatus 400 can include a processor 401 with one or more processing cores, a memory 402 with one or more computer-readable storage media, a communication component 403, and the like. The processor 401, the memory 402, and the communication component 403 are connected by a bus 404.

[0088] In the specific implementation process, the at least one processor 401 executes the computer execution instructions stored in the memory 402, so that the at least one processor 401 performs the target GNN model-based packaging design method as described above.

[0089] The specific implementation process of the processor 401 can refer to the method embodiments described above, which have similar implementation principles and technical effects, and will not be described here in detail.

[0090] Further, the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), and the like. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the present application can be directly embodied as hardware processor execution or combined execution by hardware and software modules in the processor.

[0091] The memory can include a random access memory (RAM) and can also include a non-volatile memory (NVM), such as at least one disk memory.

[0092] The bus can be an industry standard architecture (ISA) bus, a peripheral component (PCI) bus, an extended industry standard architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, the bus in the drawings of the present application does not limit to only one bus or one type of bus.​

[0093] In some embodiments, a computer program product is also provided, including a computer program or instructions, which, when executed by a processor, implement the steps of any of the above packaging design methods based on the target GNN model.

[0094] The specific implementation of the above operations can refer to the previous embodiments, which will not be repeated here.

[0095] Those skilled in the art can understand that all or part of the steps of the various methods of the above embodiments can be completed by instructions, or by relevant hardware controlled by the instructions, which can be stored in a computer readable storage medium and loaded and executed by a processor.

[0096] To this end, an embodiment of the present application provides a computer readable storage medium, which stores a plurality of program codes. The program codes can be loaded by a processor to execute the steps of any of the packaging design methods based on the target GNN model provided by the embodiments of the present application.

[0097] The storage medium can include a read only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0098] According to an aspect of the present application, a computer program product or computer program is provided, which includes computer instructions stored in a computer readable storage medium.

[0099] Due to the instructions stored in the storage medium, the steps of any of the packaging design methods based on the target GNN model provided by the embodiments of the present application can be executed, so that the beneficial effects of any of the packaging design methods based on the target GNN model provided by the embodiments of the present application can be achieved. Details are described in the previous embodiments, which will not be repeated here.

[0100] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. The specification and examples given are exemplary only and the true scope and spirit of the application is indicated by the appended claims. It will be appreciated by those skilled in the art that changes could be made to the embodiments described above, without departing from the broad discipline concept of the application. It is intended to be covered by the scope of the claims.

[0101] It should be understood that the present application is not limited to the precise structures described and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof.

Claims

1. A packaging design method based on a target GNN model, characterized in that, The method includes: Feature recognition is performed on the image to be designed to obtain each image feature and the image vector corresponding to the image feature. Based on each image vector and its corresponding target text vector, the target design vector is determined. Based on the target GNN model, the material similarity value between the target design vector and each material node vector, and the process similarity value between the target design vector and each process node vector are calculated. Based on the material similarity value and the process similarity value, multiple target design schemes are determined. The target design scheme includes target material and target process. Based on the user rating and predicted rating corresponding to each of the target design schemes, a rating loss value is calculated, and a divergence value is calculated based on preset materials, preset processes, multiple historical design schemes, and the target design scheme; the predicted rating value is a value determined based on the target GNN model. The target loss value is obtained by weighting the score loss value and the divergence value according to the preset weights. The parameters of the target GNN model are then adjusted according to the target loss value and the preset loss threshold to obtain the adjusted target GNN model, so that the target loss value calculated by the adjusted target GNN model is less than the preset loss threshold.

2. The method according to claim 1, characterized in that, The step of determining the target design vector based on each of the image vectors and their corresponding target text vectors includes: The text to be designed is semantically segmented to obtain multiple sub-texts, and the vector corresponding to each sub-text is determined to obtain multiple text vectors; Based on the CLIP model's contrastive learning mechanism, the target text vector corresponding to each image vector in the text vector is determined, and the image vector and the target text vector are associated to obtain the target vector; Based on the attention mechanism of the CLIP model, vector fusion is performed on all the target vectors and the text vectors to obtain the target design vector.

3. The method according to claim 1, characterized in that, Before calculating the material similarity value between the target design vector and each material node vector, and the process similarity value between the target design vector and each process node vector, based on the target GNN model, the method further includes: Based on the CLIP model, sample images, and corresponding sample text, the corresponding sample design vector is determined. Then, based on a preset design database, the positive and negative sample schemes corresponding to the sample design vector are determined. The positive sample scheme is the sample scheme that is associated with the sample design vector, and the negative sample scheme is the sample scheme that is not associated with the sample design vector. Determine the positive sample vector corresponding to the positive sample scheme and the negative sample vector corresponding to the negative sample scheme, and calculate the positive similarity value between the sample design vector and the positive sample vector and the negative similarity value between the sample design vector and the negative sample vector based on the initial GNN model; Based on the difference between the negative similarity value and the positive similarity value, the corresponding loss value is determined, and the model parameters of the initial GNN model are adjusted according to the loss value and a preset threshold to obtain the target GNN model.

4. The method according to claim 3, characterized in that, The step of adjusting the model parameters of the initial GNN model based on the loss value and a preset threshold to obtain the target GNN model includes: By comparing the loss value with the preset threshold, the corresponding comparison result is obtained; If the comparison result indicates that the loss value is less than the preset threshold, then the model parameters of the initial GNN model are adjusted according to the loss value to obtain the target GNN model, so that the loss value calculated according to the target GNN model is not less than the preset threshold. If the comparison result shows that the loss value is not less than the preset threshold, then the initial GNN model is determined to be the target GNN model.

5. The method according to claim 1, characterized in that, The step of determining multiple target design schemes based on the material similarity value and the process similarity value includes: According to the preset correspondence rules, at least one process node corresponding to each material node is determined, and each material node and its corresponding process nodes are determined as target design nodes; the material node is the node corresponding to the material node vector, and the process node is the node corresponding to the process node vector. The target material similarity value and target process similarity value corresponding to each target design node are added together to obtain multiple design similarity values. The design similarity values ​​are then sorted from largest to smallest to obtain a design similarity value sequence. The target material similarity value is the material similarity value corresponding to the material node in the target design node, and the target process similarity value is the process similarity value corresponding to the process node in the target design node. A preset number of design similarity values ​​in the design similarity value sequence are determined sequentially to obtain multiple target design similarity values. Based on the target material node and target process node corresponding to each target design similarity value, the corresponding target material and target process are determined. Based on each of the target materials and the target processes, multiple corresponding target design schemes are determined.

6. The method according to claim 1, characterized in that, The step of calculating the rating loss value based on the user rating value and predicted rating value corresponding to each of the target design schemes includes: Identify and output multiple target design schemes; Based on the output feedback results corresponding to the target design scheme, determine the user rating value corresponding to each target design scheme, and based on the target GNN model, determine the predicted rating value corresponding to each target design scheme; Based on the user rating and the corresponding predicted rating, the error value corresponding to each of the target design schemes is calculated, and the average value corresponding to all the error values ​​is calculated to obtain the rating loss value.

7. The method according to claim 1, characterized in that, The step of calculating the divergence value based on preset materials, preset processes, multiple historical design schemes, and the target design scheme includes: Based on a preset design database, the preset materials, the preset processes, and multiple historical design schemes are determined. Then, based on the preset materials and the preset processes, the target material scheme and the target process scheme are determined respectively from the historical design schemes and the target design scheme. The target material scheme is a scheme that includes the preset materials, and the target process scheme is a scheme that includes the preset processes. Based on the target material quantities corresponding to the target design scheme and the historical design scheme respectively, the material divergence value is calculated, and based on the target process quantities corresponding to the target design scheme and the historical design scheme respectively, the process divergence value is calculated; the target material quantity is the quantity corresponding to the target material scheme, and the target process quantity is the quantity corresponding to the target process scheme. The divergence value is obtained by weighting and summing the target material quantity and the target process quantity according to their respective preset weight values.

8. A packaging design device based on a target GNN model, characterized in that, The device includes: The recognition module is used to perform feature recognition on the image to be designed, obtain each image feature and the image vector corresponding to the image feature, and determine the target design vector based on each image vector and its corresponding target text vector; The determination module is used to calculate the material similarity value between the target design vector and each material node vector, and the process similarity value between the target design vector and each process node vector, based on the target GNN model, and to determine multiple target design schemes based on the material similarity value and the process similarity value; the target design scheme includes target material and target process; The calculation module is used to calculate the score loss value based on the user rating value and predicted rating value corresponding to each of the target design schemes, and to calculate the divergence value based on the preset materials, preset processes, multiple historical design schemes and the target design scheme; the predicted rating value is a value determined based on the target GNN model; The adjustment module is used to perform weighted calculation on the scoring loss value and the divergence value according to preset weights to obtain a target loss value, and to adjust the parameters of the target GNN model according to the target loss value and a preset loss threshold to obtain an adjusted target GNN model, so that the target loss value calculated according to the adjusted target GNN model is less than the preset loss threshold.

9. A computer device, characterized in that, include: One or more processors; Memory; One or more programs, wherein the one or more programs are stored in memory and configured to be executed by one or more processors, the one or more programs being configured to perform the 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 program code that can be called by a processor to perform the method as described in any one of claims 1 to 7.

Images