Miao batik flower and plant pattern design method, system, equipment and medium

Abstract

The invention relates to the field of computer vision, and discloses a Miao batik flower and plant pattern design method, system, equipment and medium based on a diffusion model and fuzzy comprehensive evaluation.The method comprises the steps that Miao batik flower and plant patterns are collected to construct a Miao batik flower and plant pattern material library; learning the Miao batik flower and plant pattern material library by using the diffusion model to generate an innovative flower and plant pattern conforming to a Miao batik rule; based on a preset evaluation standard, performing evaluation sorting on the innovative flower and plant patterns through a fuzzy approximation ideal solution sorting method, and screening out a flower and plant pattern set meeting satisfaction; and performing geometric rule deduction on the flower and plant pattern set through a shape grammar, and performing optimization in combination with a diffusion model to obtain a final innovative batik pattern. According to the method, stable flower and plant pattern generation quality can be ensured, and aesthetic evaluation and optimization criteria can be provided, so that a designer can select designs meeting aesthetic standards from the generated flower and plant patterns.

Description

Technical Field

[0001] The present invention relates to the field of computer vision technology, and in particular to a method, system, equipment and medium for designing Miao batik floral patterns based on a diffusion model and a fuzzy approximation ideal solution sorting method. Background Art

[0002] Under the impact of modern culture, the original charm of Miao batik has gradually faded, and it faces inheritance difficulties such as low market demand, complex craftsmanship, and reduced cultural identity. In addition, many Miao batik cultural and creative products are still limited to traditional design and application, with single floral patterns, low branding and lack of breakthrough innovation. This situation limits the promotion of Miao batik cultural and creative products in domestic and foreign markets. Therefore, in order to effectively protect and inherit this precious cultural heritage, it is necessary to combine modern technology and fashion design methods to open up new market opportunities, so as to increase the sales and cultural added value of Miao batik and promote the rebirth of Miao batik in modern society.

[0003] In recent years, the use of artificial intelligence for artistic creation has gradually become a mainstream phenomenon, and it has been fully applied in the innovative design of cultural elements such as intangible cultural heritage patterns. At present, many scholars focus on using artificial intelligence to learn the design rules of intangible cultural heritage products and automatically generate new product solutions with ethnic style. However, when dealing with complex traditional patterns, the generated patterns lack details or have blurred pixels, which cannot provide designers with accurate references.

[0004] Recently, the release of the diffusion model marks a major breakthrough in the field of intelligent image generation. However, there are still two urgent problems to be solved in the application of the diffusion model in the design of traditional floral patterns: (1) The floral patterns of the Miao ethnic group batik have rich cultural symbolic meanings, and the floral patterns generated by the algorithm alone may not be able to accurately express these traditional cultural connotations; (2) The quality of floral pattern generation is uneven, and designers lack a unified aesthetic evaluation criterion to find the best from big data.

[0005] In traditional flower and grass pattern design, the quality of flower and grass patterns generated by diffusion models varies greatly, and there is a lack of unified aesthetic evaluation criteria. It is difficult for designers to find the best design solution from big data. Therefore, it is necessary to establish a method that can not only ensure the stability of flower and grass pattern generation quality, but also provide aesthetic evaluation and optimization criteria, so that designers can select designs that meet aesthetic standards from the generated flower and grass patterns. Summary of the invention

[0006] Based on this, this application proposes a Miao batik floral pattern design method, system, equipment and medium based on diffusion model and fuzzy comprehensive evaluation, aiming to establish a set of methods that can not only ensure the stable quality of floral pattern generation, but also provide aesthetic evaluation and optimization criteria, so that designers can select designs that meet aesthetic standards from the generated floral patterns.

[0007] The first aspect of the present application provides a method for designing a Miao batik floral pattern, the method comprising: Step S1: Collecting Miao batik flower and plant patterns to build a Miao batik flower and plant pattern material library; Step S2: using a diffusion model to learn the Miao batik flower and grass pattern material library to generate innovative flower and grass patterns that conform to the rules of Miao batik; Step S3: Based on the preset evaluation criteria, the innovative flower and grass patterns are evaluated and ranked by using the fuzzy approximation ideal solution ranking method to screen out a set of flower and grass patterns that meet the satisfaction standards; Step S4: deduce geometric rules of the floral pattern set through shape grammar, and optimize it in combination with a diffusion model to obtain a final innovative batik pattern.

[0008] As an optional implementation manner of the first aspect, the step S1 includes: step S11: collecting Miao batik floral patterns from academic literature, online intangible cultural heritage databases and Miao culture-related book resources to build a systematic digital resource library; step S12: screening the floral patterns in the digital resource library, eliminating blurred, cluttered background and poor quality floral patterns, and adjusting the size of the selected floral patterns to obtain selected floral patterns; step S13: constructing a high-quality Miao batik floral pattern material library based on the selected floral patterns.

[0009] As an optional implementation of the first aspect, the step S2 includes: step S21: based on the forward diffusion process of the diffusion model, adding noise to the floral patterns in the Miao batik floral material library to obtain a pure noise image; step S22: based on the reverse diffusion process of the diffusion model, removing noise from the pure noise image to restore the detail features of the floral patterns in the Miao batik material library; step S23: learning the generation rules of Miao batik based on the detail features, and obtaining the innovative floral patterns according to the generation rules.

[0010] As an optional implementation of the first aspect, the step S3 includes: step S31: classify the innovative floral patterns to obtain multiple design schemes; step S32: preset multiple evaluation criteria based on the innovative floral patterns; step S33: construct a fuzzy decision matrix based on the design schemes and the evaluation criteria; step S34: standardize the fuzzy decision matrix to obtain a standardized fuzzy decision matrix; step S35: based on the standardized fuzzy decision matrix, determine the optimal fuzzy number and the worst fuzzy number under each evaluation criterion; step S36: calculate the first distance between each of the design schemes and the optimal fuzzy number according to the Euclidean distance formula, and calculate the second distance between each of the design schemes and the worst fuzzy number; step S37: calculate the relative proximity of the first distance and the second distance according to the Euclidean distance formula, and sort the design schemes according to the relative proximity to screen out a set of floral pattern that meets the satisfaction level.

[0011] As an optional implementation of the first aspect, in step S33: the fuzzy decision matrix is ​​expressed as , where m represents m design options, n represents n evaluation criteria, and each element in the fuzzy decision matrix represents the fuzzy score of the design under the evaluation criteria, The representation form is triangular fuzzy number , Indicates the lowest rating. Indicates the middle score, Indicates the highest rating.

[0012] As an optional implementation of the first aspect, the step S4 includes: step S41: integrating the floral pattern set into the Miao batik pattern library to obtain a first library; step S42: performing geometric rule deduction and design on the floral patterns in the first library through shape grammar to form a second library; step S43: importing the second library into the Midjourney software, and entering keywords to obtain a new Miao batik pattern; step S44: using the new Miao batik pattern as an optimized data set for the diffusion model to obtain the final innovative batik pattern.

[0013] As an optional implementation of the first aspect, the shape grammar in step S42 includes general rules and adjustment rules, the general rules include replacement and addition and deletion, and the adjustment rules include proportion, copy, mirror, crop, rotation and coordination.

[0014] A second aspect of the present application provides a Miao batik pattern design system, the system comprising: The pattern collection and storage module is used to collect the Miao batik flower and grass patterns to build a Miao batik flower and grass pattern material library; A diffusion model pattern generation module is used to use the diffusion model to learn the Miao batik flower and grass pattern material library to generate innovative flower and grass patterns that conform to the rules of Miao batik; A multi-dimensional evaluation and screening module is used to evaluate and sort the innovative flower and grass patterns based on preset evaluation criteria by using a fuzzy approximation ideal solution sorting method, and screen out a set of flower and grass patterns that meet satisfaction standards; The rule deduction and optimization module is used to deduce the geometric rules of the flower and grass pattern set through shape grammar, and optimize it in combination with the diffusion model to obtain the final innovative batik pattern.

[0015] The third aspect of the present application provides an electronic device, comprising: a processor; a memory for storing executable instructions of the processor; wherein the processor is configured to execute the executable instructions to implement the above-mentioned Miao batik floral pattern design method.

[0016] A fourth aspect of the present application provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the above-mentioned Miao batik floral pattern design method.

[0017] Compared with the prior art, the present application provides a method for designing Miao batik floral patterns: first, a diffusion model is used to learn the Miao batik floral pattern material library. Through the diffusion model, the dyeing and diffusion process of Miao batik can be simulated, so as to generate innovative floral patterns that conform to the rules of Miao batik, and realize the essence of traditional Miao batik art and innovation combined with modern technology; secondly, the innovative floral patterns are evaluated and ranked by the fuzzy approximation ideal solution sorting method, which not only ensures that the designed floral patterns are innovative, but also meets the preset evaluation criteria. The fuzzy approximation ideal solution sorting method can handle fuzzy and uncertain information, making the evaluation results more accurate and comprehensive; then, the selected floral pattern set is designed by geometric rule deduction through shape grammar, and optimized in combination with the diffusion model. The shape grammar can provide a systematic and regular method for shape change and design of the pattern, so that the innovative floral patterns can be further optimized and evolved on the basis of maintaining the original characteristics. Finally, combined with the optimization of the diffusion model, the details and overall effect of the floral patterns can be further adjusted and optimized.

[0018] Additional aspects and advantages of the present application will be given in part in the following description, and in part will become apparent from the following description, or will be understood through the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flowchart of a method for designing a Miao batik floral pattern proposed in the first embodiment of the present application; Figure 2 This is a schematic structural diagram of a Miao batik floral pattern design system proposed in the second embodiment of the present application.

[0020] The following specific implementation methods will further illustrate the present application in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here. In addition, the "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally represents that the objects associated with each other are in an "or" relationship.

[0023] In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.

[0024] Example 1 See also Figure 1 , which is a flow chart of a method for designing Miao batik floral patterns proposed in the first embodiment of the present application, wherein the proposed method includes S1 to S4.

[0025] Step S1: Collect Miao batik floral patterns to build a Miao batik floral pattern material library.

[0026] In some implementations, step S1 specifically includes the following sub-steps.

[0027] Step S11: Collecting Miao batik floral patterns from academic literature, online intangible cultural heritage databases, and Miao culture-related book resources to build a systematic digital resource library; Step S12: screening the flower and grass patterns in the digital resource library, removing blurred, cluttered background and poor quality flower and grass patterns, and resizing the selected flower and grass patterns to obtain selected flower and grass patterns; Step S13: constructing a high-quality Miao batik floral pattern material library based on the selected floral patterns.

[0028] In the implementation method of this step, by collecting academic literature, online intangible cultural heritage databases and Miao culture-related book resources, a systematic digital resource library is constructed, which can widely and comprehensively obtain various types of Miao batik flower and grass pattern patterns, providing a rich data foundation for subsequent work. The flower and grass pattern patterns in the digital resource library are screened, and the blurred, cluttered background and poor quality flower and grass pattern patterns are eliminated, and the size of the selected patterns is adjusted, which can effectively improve the quality and standardization of the flower and grass pattern patterns and ensure the high level of the patterns in the material library. Finally, a high-quality Miao batik flower and grass pattern material library is constructed, which provides high-quality and standardized material support for the subsequent research, application, inheritance and innovation of Miao batik flower and grass patterns.

[0029] Step S2: Use the diffusion model to learn the Miao batik floral pattern library to generate innovative floral patterns that conform to the rules of Miao batik.

[0030] It should be noted that the underlying logical architecture of the diffusion model includes the forward diffusion process and the reverse diffusion process, which learns the data distribution by adding and removing noise to generate high-quality data. The reverse diffusion process starts with Gaussian noise and generates data with a distribution similar to the original data after T steps of reverse diffusion.

[0031] In some implementations, step S2 specifically includes the following sub-steps.

[0032] Step S21: Based on the forward diffusion process of the diffusion model, noise is added to the flower and plant patterns in the Miao batik flower and plant material library to obtain a pure noise image.

[0033] Specifically, during the forward diffusion process, the model gradually transforms the original floral pattern into pure noise by adding noise in multiple steps. Each step adds a small amount of noise according to a certain probability distribution to ensure that the final noisy image still contains some information of the original image. Specifically, given an original data, the forward diffusion process is usually described by a conditional probability distribution. After T steps of noise addition, we get a Gaussian noise close to The forward diffusion process can also be viewed as a series of steps that gradually add noise, similar to the diffusion of ink in water.

[0034] Among them, the expression of the conditional probability distribution of the forward diffusion process is: , In the formula, Indicates that in a given Under the conditions The conditional probability distribution of ; represents a normal distribution (Gaussian distribution) used to simulate and control the noise addition and removal steps in the process of generating batik flower pattern images; is the data at time step t; is the mean of the distribution, indicating that the data at time step t is the data at the previous time step The data has passed a certain proportion Attenuation of the original flower and grass pattern image, while adding noise, also retains the outline and shape of some of the original flower and grass pattern images; Determines how much information of the original floral pattern image is retained. The larger it is, the less image information is retained and the more noise is generated; is the covariance matrix, which determines the intensity of the noise added to the floral pattern image. Repeat this process T steps until the image Close to pure noise, this process is similar to ink spreading in water, gradually blurring the image.

[0035] Step S22: Based on the reverse diffusion process of the diffusion model, the pure noise image is de-noised to restore the detailed features of the flower and plant patterns in the Miao batik material library.

[0036] Specifically, in the reverse diffusion process, the model performs the opposite operation to the forward process, starting from Gaussian noise, and after T steps of reverse diffusion, gradually removes the noise, reveals the features and details hidden under the noise, and generates high-quality generated samples; the conditional probability distribution of the reverse diffusion process requires model learning, which is usually parameterized as a neural network such as U-Net, whose input is the noise image of the current time step and time step T, the output is the image of the previous time step prediction; through this process, the diffusion model is combined with the LoRA model to generate batik floral patterns.

[0037] Among them, the expression of the conditional probability distribution of the reverse diffusion process is: , In the formula, is the output of the neural network U-Net, predicting the batik flower pattern image of the previous time step, is the covariance matrix, which is used to control the intensity of the noise.

[0038] Step S23: learning the generation rules of Miao batik according to the detail features, and obtaining innovative floral patterns according to the generation rules.

[0039] In the implementation method of this step, noise is added to the floral pattern in the Miao batik material library through the forward diffusion process to obtain a pure noise image, which helps to better restore and reveal the detailed features of the floral pattern in the subsequent reverse diffusion process, and provide rich detail information for innovative design. Based on the reverse diffusion process of the diffusion model, the pure noise image is de-noised, and the detailed features of the floral pattern in the Miao batik material library are restored. The reverse generation capability of the diffusion model is fully utilized, and high-quality generation samples can be generated, retaining the features and details of the original floral pattern. The generation rules of Miao batik are learned based on the restored detailed features, and innovative floral patterns are obtained based on these rules, realizing the transformation from the learned rules to innovative design. Through the powerful generation capability of the diffusion model, innovative floral patterns that conform to the laws of Miao batik art are generated.

[0040] Step S3: Based on the preset evaluation criteria, the innovative flower and grass patterns are evaluated and ranked by the fuzzy approximation ideal solution ranking method to screen out a set of flower and grass patterns that meet the satisfaction standards.

[0041] In some implementations, step S3 specifically includes the following sub-steps.

[0042] Step S31: classify the innovative floral patterns to obtain a variety of design solutions.

[0043] Step S32: Preset multiple evaluation criteria based on the innovative floral patterns.

[0044] Step S33: Construct a fuzzy decision matrix according to the design scheme and evaluation criteria.

[0045] Specifically, suppose there are m design solutions , n evaluation criteria , the evaluator scores each plan according to these evaluation criteria to form a fuzzy decision matrix , each element in the fuzzy decision matrix D It is the fuzzy score of the design scheme under the evaluation criteria, usually expressed as a triangular fuzzy number . Among them, the m design schemes refer to the m different innovative floral patterns generated by the diffusion model. These innovative floral patterns are candidate schemes in the design process and need to be further evaluated for their market acceptance and cultural value. Exemplarily, the evaluation criteria B1, B2, B3, B4 and B5 represent symmetry, balance, strong sense of rhythm, harmonious proportion and overall unity, respectively. The Likert Seven-Point Scale is set from 1 to 7. Respondents choose an option that best represents their opinions based on their true feelings, and fuzzify customer evaluations through language variables. The scoring results are expressed in the form of triangular fuzzy numbers: ,in is the lowest possible score, is the most likely middle score; is the highest possible score.

[0046] Step S34: Standardize the fuzzy decision matrix to obtain a standardized fuzzy decision matrix.

[0047] Specifically, the min-max normalization formula is used to normalize the fuzzy decision matrix. The formula is: , where Representative batik pattern design In the evaluation criteria The standardized score under Represents all programs in the evaluation criteria The lowest rating under Represents all programs in the evaluation criteria The highest rating under . It can refer to any of the five evaluation indicators. Through standardization, the score of each design scheme under each evaluation criterion is normalized to the interval [0,1], which is convenient for subsequent comparison and calculation.

[0048] Step S35: Based on the standardized fuzzy decision matrix, determine the optimal fuzzy number and the worst fuzzy number under each evaluation criterion.

[0049] Specifically, using the fuzzy positive ideal solution and fuzzy negative ideal solution They represent the optimal fuzzy number and the worst fuzzy number under each evaluation criterion respectively.

[0050] For positive ideal solutions: ; For negative ideal solutions: ; In the formula, It represents the optimal fuzzy score set under all evaluation criteria, which represents the best performance index that can be achieved by each evaluation criterion when designing Miao batik floral patterns; represents the optimal fuzzy score under the first evaluation criterion "symmetry", which means the best score among all batik flower pattern designs under this evaluation criterion. And so on. It represents the worst fuzzy score set under all evaluation criteria, representing the worst performance index that can be achieved by each evaluation criterion when designing Miao batik floral patterns; It is the worst fuzzy score under the first evaluation criterion "Symmetry", indicating the worst score among all batik flower pattern designs under this criterion. And so on.

[0051] Step S36: Calculate the first distance between each design solution and the optimal fuzzy number, and calculate the second distance between each design solution and the worst fuzzy number according to the Euclidean distance formula.

[0052] Specifically, the Euclidean distance formula is used to calculate the distance between each design solution and the fuzzy positive ideal solution. and negative ideal solution The distance between.

[0053] The distance calculation formula to the positive ideal solution is: , In the formula, It represents the optimal fuzzy value under each evaluation criterion, that is, the most ideal batik floral pattern design. Indicates the sum of all n evaluation criteria; Indicates i The batik floral pattern design is in j The scores on the evaluation criteria represent the lowest score, the middle score, and the highest score respectively. The optimal batik flower pattern design is j The scores on the evaluation criteria are the lowest score, the middle score, and the highest score. Then the difference between the score on each evaluation criterion and the optimal score is calculated, and then these differences are added together to get the first distance.

[0054] The distance calculation formula to the negative ideal solution is: , In the formula, Represents the worst fuzzy value under each evaluation criterion, that is, the most undesirable batik flower pattern design; Indicates i Batik floral pattern design j The scores on the three criteria represent the lowest score, the middle score, and the highest score respectively. The worst batik floral pattern design is j The scores on each standard are also the lowest score, the middle score and the highest score. Similarly, the difference between the score on each standard and the worst score is calculated, and then these differences are added together to get the second distance.

[0055] Step S37: Calculate the relative proximity of the first distance and the second distance according to the Euclidean distance formula, sort the design solutions according to the relative proximity, and select a satisfactory flower and grass pattern set.

[0056] Specifically, the relative proximity of each batik flower pattern design is calculated and the design schemes are ranked, and the calculation formula is: , In the formula, Representativei The relative closeness of a batik floral pattern design, which is a value between 0 and 1. The closer the value is to 1, the closer the batik flower and grass pattern design is to the ideal design; the closer the value is to 0, the closer the batik flower and grass pattern design is to the worst design. It is i The second distance between the batik flower and grass pattern design and the worst design (negative ideal solution). The smaller the distance, the better the batik flower and grass pattern design. It is i The first distance between a batik flower and grass pattern design and the ideal design (positive ideal solution). The larger the distance, the closer the batik flower and grass pattern design is to the ideal. Indicates i The total distance between the batik flower pattern design and the ideal design and the worst design is used as a denominator to calculate the standardized relative proximity.

[0057] In the implementation of this step, the Likert seven-point scale is used to convert subjective evaluations (such as "symmetry" and "strong sense of rhythm") into language variables, and then mapped into triangular fuzzy numbers (lowest score, most likely score, highest score), which effectively captures the evaluator's fuzzy cognition and uncertainty. For example, the respondents' score for "proportional harmony" may be between 5-7 points, and the triangular fuzzy number (5, 6, 7) is more realistic than a single numerical value. Furthermore, by calculating the Euclidean distance between each design scheme and the positive ideal solution and the negative ideal solution, it not only measures the gap between the scheme and the ideal state, but also takes into account the degree to which it is away from the worst state. For example, a certain floral pattern Small (close to ideal) and Larger (farther away from the worst), its relative closeness The closer it is to 1, the higher it is ranked. Relative proximity combines positive and negative distances to avoid the one-sidedness of a single distance. For example, when a design scheme is excellent in one evaluation criterion but mediocre in other indicators, it may be due to and It is balanced and not overestimated, reducing the misjudgment of "biased" programs.

[0058] Step S4: The floral pattern set is deduced through geometric rules using shape grammar, and optimized in combination with the diffusion model to obtain the final innovative batik floral pattern.

[0059] In some implementations, step S4 specifically includes the following sub-steps.

[0060] Step S41: Integrate the flower and grass pattern set into the Miao batik pattern material library to obtain a first material library; Step S42: using shape grammar to deduce and design the floral patterns in the first material library according to geometric rules to form a second material library; Step S43: importing the second material library into the Midjourney software and inputting keywords to obtain a new Miao batik pattern; Step S44: using the new Miao batik pattern as an optimized data set for the diffusion model to obtain the final innovative batik pattern.

[0061] Specifically, the basic layout structure is extracted from the traditional Miao batik pattern material library. Batik flower and grass patterns are classified according to different geometric forms such as triangles, squares and circles, and the common layout structures of Miao batik flower and grass patterns such as "nine-square grid" and "rice grid" can be summarized. The structure of Miao batik patterns is extracted and summarized and numbered into the Guizhou batik database, so as to provide reference data and creative basis for subsequent innovative pattern generation.

[0062] The basic constituent elements and basic floral patterns of the Miao batik patterns are defined based on the first material library. These elements constitute the basic grammatical rules of the shape grammar. The basic floral patterns are composed into novel floral patterns according to the deduction rules of rotation, superposition and translation. And based on the structure of "nine-square grid" or "rice grid", the pattern set is placed in the traditional Miao batik framework. The flexibility of the rules is used to explore new ways of combining and composing Miao batik patterns. Exemplarily, the shape grammar includes general rules and adjustment rules. The general rules include replacement and addition and deletion, and the adjustment rules include proportion, replication, mirroring, cropping, rotation and coordination.

[0063] Based on the deduction of batik patterns using shape grammar, keywords such as "Hmong batik", "geometric flowers and plants", "symmetry", etc. are input into the artificial intelligence image generation platform Midjourney, and parameters such as similarity and size are set to quickly generate innovative batik patterns. The target pattern is selected and preprocessed to obtain a data set. After screening and cropping, the denoising and random generation mechanism of the diffusion model is used to enrich the details and variability of the pattern, and the final innovative batik pattern is obtained after human-machine collaborative innovation.

[0064] In terms of application, the traditional aesthetics of Guizhou batik can be combined with modern design concepts, and the innovative batik patterns can be finally deduced and applied to the design of women's handbags by combining shape grammar with artificial intelligence technology, as the main design element of batik women's handbags. In order to better show the overall picture of the pattern when designing the sketch, for example, the bag type is a simple and elegant tote handbag to highlight the visual effect of the batik pattern, and Chinese elements such as bamboo and mahogany are incorporated into the details such as the bag handle, which not only echoes the Chinese batik process, but also takes into account environmental protection and durability. In the process of making the sample bag, the craftsman first draws the designed pattern on the base cloth and applies wax, and then the craftsman dyes the cloth. The dye will penetrate into the uncoated area of ​​the wax, while the wax flower part remains the original color, forming a unique batik effect.

[0065] In the implementation of this step, the geometric rules of the patterns in the first material library are deduced by shape grammar to design a new Miao batik pattern, which not only retains the traditional characteristics of Miao batik, but also generates new patterns through innovative geometric rule deduction, enriching the diversity of design. Midjourney software is used to select target patterns according to keywords, and pre-process them to obtain data sets, realizing the combination of artificial intelligence and design, and improving design efficiency and accuracy; by setting keywords and parameters, innovative batik patterns that meet the requirements can be quickly generated, greatly shortening the design cycle. Through the application of the diffusion model, the details and variability of the pattern are enriched, improving the artistry and practicality of the design; at the same time, the optimization mechanism of the diffusion model can also make the innovative batik pattern more in line with the artistic characteristics and aesthetic requirements of Miao batik.

[0066] Example 2 See also Figure 2 , which is a schematic diagram of the structure of a Miao batik floral pattern design system proposed in the second embodiment of the present application, the system comprises: The pattern collection and storage module 100 is used to collect Miao batik flower and grass patterns to build a Miao batik flower and grass pattern material library; The diffusion model pattern generation module 200 is used to learn the Miao batik flower and grass pattern material library by using the diffusion model to generate innovative flower and grass patterns that conform to the rules of Miao batik; The multi-dimensional evaluation and screening module 300 is used to evaluate and sort the innovative flower and grass patterns based on preset evaluation criteria by using a fuzzy approximation ideal solution sorting method, and screen out a set of flower and grass patterns that meet satisfaction standards; The rule deduction and optimization module 400 is used to deduce geometric rules of the flower and grass pattern set through shape grammar, and optimize it in combination with the diffusion model to obtain the final innovative batik pattern.

[0067] On the other hand, the present application also proposes an electronic device, comprising: a processor; a memory for storing executable instructions of the processor; wherein the processor is configured to execute the executable instructions to implement the above-mentioned Miao batik floral pattern design method.

[0068] On the other hand, the present application also proposes a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is able to execute the above-mentioned Miao batik floral pattern design method.

[0069] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0070] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0071] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A method for designing a Miao batik floral pattern, characterized in that: The method comprises: Step S1: Collecting Miao batik flower and plant patterns to build a Miao batik flower and plant pattern material library; Step S2: using a diffusion model to learn the Miao batik flower and grass pattern material library to generate innovative flower and grass patterns that conform to the rules of Miao batik; Step S3: Based on the preset evaluation criteria, the innovative flower and grass patterns are evaluated and ranked by using the fuzzy approximation ideal solution ranking method to screen out a set of flower and grass patterns that meet the satisfaction standards; Step S4: deduce geometric rules of the floral pattern set through shape grammar, and optimize it in combination with a diffusion model to obtain a final innovative batik pattern.

2. The method for designing a Miao batik floral pattern according to claim 1, characterized in that: The step S1 comprises: Step S11: Collecting Miao batik floral patterns from academic literature, online intangible cultural heritage databases, and Miao culture-related book resources to build a systematic digital resource library; Step S12: screening the flower and grass patterns in the digital resource library, removing blurred, cluttered background and poor quality flower and grass patterns, and resizing the selected flower and grass patterns to obtain selected flower and grass patterns; Step S13: constructing a high-quality Miao batik floral pattern material library based on the selected floral patterns.

3. The method for designing a Miao batik floral pattern according to claim 1, characterized in that: The step S2 comprises: Step S21: based on the forward diffusion process of the diffusion model, adding noise to the flower and plant patterns in the Miao batik flower and plant material library to obtain a pure noise image; Step S22: based on the reverse diffusion process of the diffusion model, removing noise from the pure noise image to restore the detail features of the floral patterns in the Miao batik material library; Step S23: learning the generation rules of Miao batik according to the detail features, and obtaining the innovative floral pattern according to the generation rules.

4. The method for designing a Miao batik floral pattern according to claim 1, characterized in that: The step S3 comprises: Step S31: classifying the innovative flower and grass patterns to obtain a variety of design solutions; Step S32: presetting a plurality of evaluation criteria based on the innovative floral pattern; Step S33: constructing a fuzzy decision matrix according to the design scheme and the evaluation criteria; Step S34: normalizing the fuzzy decision matrix to obtain a standardized fuzzy decision matrix; Step S35: determining the optimal fuzzy number and the worst fuzzy number under each of the evaluation criteria based on the standardized fuzzy decision matrix; Step S36: calculating a first distance between each of the design solutions and the optimal fuzzy number, and calculating a second distance between each of the design solutions and the worst fuzzy number according to the Euclidean distance formula; Step S37: calculating the relative proximity between the first distance and the second distance according to the Euclidean distance formula, and sorting the design schemes according to the relative proximity to screen out a satisfactory flower and grass pattern set.

5. The method for designing a Miao batik floral pattern according to claim 1, characterized in that: In step S33: The fuzzy decision matrix is ​​expressed as , where m represents m design options, n represents n evaluation criteria, and each element in the fuzzy decision matrix represents the fuzzy score of the design under the evaluation criteria, The representation form is triangular fuzzy number , Indicates the lowest rating. represents the middle score, Indicates the highest rating.

6. The method for designing a Miao batik floral pattern according to claim 1, characterized in that: The step S4 comprises: Step S41: integrating the flower and grass pattern set into the Miao batik pattern material library to obtain a first material library; Step S42: using shape grammar to deduce and design geometric rules for the flower and grass patterns in the first material library to form a second material library; Step S43: importing the second material library into the Midjourney software, and inputting keywords to obtain a new Miao batik pattern; Step S44: using the new Miao batik pattern as an optimized data set for a diffusion model to obtain a final innovative batik pattern.

7. A method for designing Miao batik floral patterns according to claim 6, characterized in that: The shape grammar in step S42 includes general rules and adjustment rules, the general rules include replacement and addition and deletion, and the adjustment rules include proportion, copy, mirror, crop, rotation and coordination.

8. A Miao batik floral pattern design system, characterized in that: The system comprises: The pattern collection and storage module is used to collect the Miao batik flower and grass patterns to build a Miao batik flower and grass pattern material library; A diffusion model pattern generation module is used to use the diffusion model to learn the Miao batik flower and grass pattern material library to generate innovative flower and grass patterns that conform to the rules of Miao batik; A multi-dimensional evaluation and screening module is used to evaluate and sort the innovative flower and grass patterns based on preset evaluation criteria by using a fuzzy approximation ideal solution sorting method, and screen out a set of flower and grass patterns that meet satisfaction standards; The rule deduction and optimization module is used to deduce the geometric rules of the flower and grass pattern set through shape grammar, and optimize it in combination with the diffusion model to obtain the final innovative batik pattern.

9. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the executable instructions to implement a Miao batik floral pattern design method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method for designing Miao batik floral patterns as described in any one of claims 1 to 7.

Images