Clothing fashion trend forecasting and intelligent design platform

Abstract

The present invention discloses a clothing fashion trend prediction and intelligent design platform, comprising: a data acquisition module for acquiring raw data related to clothing; a data processing module for preprocessing and performing variational mode decomposition on the collected raw data to extract effective features from the data; a data analysis and prediction module for using a gated recurrent unit model to predict fashion trends on the processed data; an intelligent design module for recommending design elements based on the fashion trend prediction results and generating clothing design drawings using a generative adversarial network; and a platform interface module for providing a user visual interface, displaying trend analysis results and design solutions, and allowing users to make design adjustments. The clothing fashion trend prediction and intelligent design platform of the present invention effectively improves the accuracy of fashion trend prediction and the efficiency of intelligent design, reduces design costs, shortens product launch cycles, and enhances the market competitiveness of brands.

Description

Technical Field

[0001] The present invention belongs to the technical field of clothing design and manufacturing, and in particular relates to a clothing fashion trend prediction and intelligent design platform. Background Art

[0002] Against the backdrop of rapid globalization and informatization, the apparel industry faces unprecedented challenges and opportunities. Consumer demands are becoming increasingly diverse and personalized, making it difficult for traditional apparel design and manufacturing models to quickly respond to market changes, putting companies at a competitive disadvantage. Therefore, leveraging advanced technologies to improve design efficiency, predict market trends, and shorten product time-to-market have become pressing challenges for the apparel industry.

[0003] Currently, fashion trend forecasting relies primarily on designers' experience and market research. However, this approach suffers from subjectivity, incomplete data, and long forecasting cycles, making it difficult to quickly and accurately grasp market trends. Furthermore, the traditional fashion design process is often time-consuming and costly, making it difficult to guarantee the creativity and market adaptability of design solutions.

[0004] To overcome these challenges, big data and artificial intelligence technologies have been widely applied in the apparel industry in recent years. Despite this, existing technologies still have numerous shortcomings in data processing, model selection, and system integration. In particular, traditional statistical methods and simple machine learning models are unable to fully tap into the deep insights within time series data, leaving room for improvement in prediction accuracy. Furthermore, effectively integrating prediction results with apparel design to achieve the automated generation and optimization of design solutions remains a hot topic and a challenge. Summary of the Invention

[0005] The present invention provides a clothing fashion trend prediction and intelligent design platform to solve the above-mentioned technical problems, specifically adopting the following technical solutions:

[0006] A clothing fashion trend prediction and intelligent design platform, including:

[0007] Data acquisition module, used to obtain raw data related to clothing;

[0008] The data processing module is used to pre-process and perform variational mode decomposition on the collected raw data and extract effective features from the data;

[0009] Data analysis and prediction module, used to predict the popularity trend of processed data using the gated recurrent unit model;

[0010] An intelligent design module that recommends design elements based on trend predictions and generates clothing designs using a generative adversarial network.

[0011] The platform interface module is used to provide a user visual interface, display trend analysis results and design solutions, and allow users to make design adjustments.

[0012] Furthermore, the data acquisition module performs data acquisition through the following steps:

[0013] Collect sales data from e-commerce platforms, including product sales volume, sales revenue, and best-selling products;

[0014] Collecting review data from user review areas, including user evaluations, ratings, and review content of products;

[0015] Collect discussion data from social media platforms, including topics, tags, post content, and user interaction data related to clothing;

[0016] The press conference information is collected from the fashion press conference, and the press conference information includes the press conference time, brand, display clothing style, color, and material.

[0017] Furthermore, the data collection module regularly collects data through web crawler technology and API interface, and stores the data in the database.

[0018] Furthermore, the data processing module includes:

[0019] The data cleaning unit is used to remove noise data, duplicate data, and incomplete data from the collected raw data, unify the data format, and fill in missing values;

[0020] The variational mode decomposition unit is used to perform multi-scale decomposition on the cleaned data and extract various modal components.

[0021] Furthermore, the formula of the variational mode decomposition is:

[0022] ;

[0023] in, For the modal, is its center frequency.

[0024] Furthermore, the gated recurrent unit model includes:

[0025] Update gate, used to control the impact of the previous state on the current state;

[0026] Reset gate, used to control the impact of current input on the current state;

[0027] The current state unit is used to calculate the state at the current moment.

[0028] Furthermore, the mathematical formula of the gated recurrent unit model is:

[0029] ;

[0030] ;

[0031] ;

[0032] ;

[0033] in, To update the gate, To reset the gate, is the candidate state at the current moment, is the state at the current moment, W and b are model parameters.

[0034] Furthermore, the intelligent design module includes:

[0035] Design element library, used to store a database containing color, style, material, and pattern design elements;

[0036] A design element recommendation unit is used to select elements that meet the predicted trend from the design element library based on the trend prediction results;

[0037] Generative adversarial networks are used to generate clothing designs based on recommended design elements.

[0038] Furthermore, the mathematical formula of the generative adversarial network is:

[0039] ;

[0040] Among them, D is the discriminator, G is the generator, is the real data sample, z is the noise sample, is the distribution of real data, is the distribution of noise samples.

[0041] Furthermore, the platform interface module includes:

[0042] Trend analysis result display unit, used to visually display the results of popular trend prediction;

[0043] Design scheme display unit, used to display automatically generated clothing design schemes for users to preview, edit and select the final design;

[0044] The design adjustment unit is used to provide online design tools for users to fine-tune and customize the design plan according to their needs.

[0045] The benefits of this invention lie in the provided clothing fashion trend forecasting and intelligent design platform, which enables high-precision forecasting of clothing fashion trends and intelligent design generation, significantly improving design efficiency and market response speed, reducing design costs and error risks, and enhancing the market competitiveness and brand influence of enterprises. Specifically,

[0046] 1. Improve the accuracy of trend prediction:

[0047] Multi-source data integration: The platform obtains data from multiple channels such as e-commerce platforms, user reviews, social media, and fashion shows. The data sources are extensive and rich, ensuring the comprehensiveness and representativeness of the data.

[0048] Advanced time series analysis: Using variational mode decomposition (VMD) to perform multi-scale decomposition of time series data, extract different modal components, effectively capture the complex features in the data, and improve the accuracy of data processing and feature extraction.

[0049] Deep learning prediction model: Using the gated recurrent unit (GRU) model, it effectively utilizes historical information and current input through gate update and reset mechanisms, improving the accuracy and robustness of trend prediction.

[0050] 2. Realize intelligent design:

[0051] Automated design element recommendation: Based on the results of popular trend predictions, intelligent recommendations are made from the design element library on colors, styles, materials, patterns and other design elements that conform to current popular trends, reducing the designer's workload and time cost.

[0052] Generative Adversarial Network (GAN): GAN technology is used to generate clothing design drawings that conform to recommended design elements, ensuring the novelty and diversity of design solutions while retaining the creativity and market adaptability of the designs.

[0053] 3. Improve design efficiency and market response speed:

[0054] Full-process automation: The platform realizes full-process automation from data collection, processing, analysis to design generation, greatly improving work efficiency and shortening the product design and market launch cycle.

[0055] User participation and adjustment: A visual interface is provided to display trend analysis results and design solutions. Users can preview, edit and select the final design solution online, and use design adjustment tools to fine-tune and customize it, thereby improving user satisfaction and the market adaptability of the design solution.

[0056] 4. Reduce costs and risks:

[0057] Reduce manual intervention: Through automated data processing and design generation, the reliance on designer experience and manual operations is reduced, which reduces labor costs and the risk of errors.

[0058] Improve market competitiveness: Accurate trend forecasts and efficient intelligent design enable companies to quickly respond to market changes, launch products that meet market demand, and enhance the brand's market competitiveness.

[0059] 5. Significant technological advantages:

[0060] Combination of VMD and GRU: The combination of variational mode decomposition (VMD) and gated recurrent unit (GRU) fully utilizes the time series decomposition capability of VMD and the deep learning prediction capability of GRU to provide accurate popular trend prediction.

[0061] GAN generative design: The application of generative adversarial networks (GANs) in clothing design ensures the innovation and diversity of design schemes while taking into account market demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0063] Figure 1 This is a flow chart of a clothing fashion trend prediction and intelligent design platform of the present invention. DETAILED DESCRIPTION

[0064] The following describes in detail embodiments of the present application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0065] This application discloses a clothing fashion trend prediction and intelligent design platform. The flowchart of the clothing fashion trend prediction and intelligent design platform is as follows: Figure 1 As shown. The clothing fashion trend forecasting and intelligent design platform includes: data acquisition module, data processing module, data analysis and forecasting module and platform interface module. Figure 1 The flowchart introduces the various modules of the clothing fashion trend prediction and intelligent design platform of this application.

[0066] The data acquisition module is used to obtain raw data related to clothing. In the embodiment of the present application, the data acquisition module regularly collects data through web crawler technology and API interface and stores it in the database.

[0067] Specifically, web crawlers are used to regularly collect data from e-commerce and social media platforms. The crawlers simulate user visits to the websites and automatically extract the required data. Structured data is obtained from various data sources through APIs. Many e-commerce and social media platforms offer APIs that allow developers to access data from the platforms. All data is stored in a centralized database for subsequent processing and analysis.

[0068] In the embodiment of the present application, the data acquisition module performs data acquisition through the following steps:

[0069] Collect sales data from e-commerce platforms, including product sales volume, sales revenue, and hot-selling products.

[0070] Collect review data from the user review area, which includes user evaluation, rating, and review content of the product.

[0071] Discussion data is collected from social media platforms, including clothing-related topics, tags, post content, and user interaction data.

[0072] Collect press conference information from fashion shows, including press conference time, brand, display clothing style, color, and material.

[0073] The data processing module is used to preprocess and perform variational mode decomposition on the collected raw data and extract effective features from the data.

[0074] In an embodiment of the present application, the data processing module includes:

[0075] The data cleaning unit is used to remove noise data, duplicate data, and incomplete data from the collected raw data, unify the data format, and fill in missing values.

[0076] The variational mode decomposition (VMD) unit is used to perform multi-scale decomposition on the cleaned data (time series data) and extract the various modal components. VMD can decompose complex time series into several intrinsic mode functions (IMFs), each of which represents a different frequency component, thereby better capturing the characteristics of the data and ultimately outputting the processed data features ( ).

[0077] In the embodiment of the present application, the formula of variational mode decomposition is:

[0078] ;

[0079] in, For the modal, is its center frequency.

[0080] The data analysis and prediction module is used to predict popular trends based on processed data using the Gated Recurrent Unit (GRU) model.

[0081] In an embodiment of the present application, the gated recurrent unit model includes:

[0082] Update gate, used to control the impact of the previous state on the current state.

[0083] Reset gate, used to control the impact of current input on the current state.

[0084] The current state unit is used to calculate the state at the current moment.

[0085] In the embodiment of the present application, the mathematical formula of the gated recurrent unit model is:

[0086] ;

[0087] ;

[0088] ;

[0089] ;

[0090] in, To update the gate, To reset the gate, is the candidate state at the current moment, is the state at the current moment, W and b are model parameters.

[0091] The intelligent design module is used to recommend design elements based on fashion trend prediction results and generate clothing design drawings using a generative adversarial network.

[0092] In an embodiment of the present application, the intelligent design module includes:

[0093] The design element library is used to store a database of design elements including colors, styles, materials, and patterns. Each design element is labeled with a corresponding attribute to facilitate the recommendation system to filter and match.

[0094] The design element recommendation unit is used to select elements from the design element library that match the predicted trends based on trend prediction results. The recommendation system selects matching design elements based on key features of the trend prediction results (such as popular colors, styles, materials, etc.).

[0095] Generative Adversarial Networks (GANs) are used to generate clothing designs based on recommended design elements.

[0096] In the embodiment of the present application, the mathematical formula for generating an adversarial network is:

[0097] ;

[0098] Among them, D is the discriminator, G is the generator, is the real data sample, z is the noise sample, is the distribution of real data, is the distribution of noise samples.

[0099] The platform interface module is used to provide a user visual interface, display trend analysis results and design solutions, and allow users to make design adjustments.

[0100] In an embodiment of the present application, the platform interface module includes:

[0101] The trend analysis result display unit is used to visually display the results of popular trend forecasts. Users can intuitively understand market trends and future trends through charts, statistical data, etc.

[0102] The design scheme display unit is used to display automatically generated clothing design schemes for users to preview, edit and select the final design.

[0103] The design adjustment unit is used to provide online design tools for users to fine-tune and customize the design plan according to their needs.

[0104] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the above embodiments do not limit the present invention in any form, and any technical solutions obtained by equivalent replacement or equivalent transformation fall within the scope of protection of the present invention.

Claims

1. A clothing fashion trend prediction and intelligent design platform, characterized by: Include: Data acquisition module, used to obtain raw data related to clothing; The data processing module is used to pre-process and perform variational mode decomposition on the collected raw data and extract effective features from the data; Data analysis and prediction module, used to predict the popularity trend of processed data using the gated recurrent unit model; An intelligent design module that recommends design elements based on trend predictions and generates clothing designs using a generative adversarial network. The intelligent design module includes: Design element library, used to store a database containing color, style, material, and pattern design elements; A design element recommendation unit is used to select elements that meet the predicted trend from the design element library based on the trend prediction results; Generative adversarial networks, used to generate clothing designs based on recommended design elements; The platform interface module is used to provide a user visual interface, display trend analysis results and design solutions, and allow users to make design adjustments; The platform interface module includes: Trend analysis result display unit, used to visually display the results of popular trend prediction; Design scheme display unit, used to display automatically generated clothing design schemes for users to preview, edit and select the final design; The design adjustment unit is used to provide online design tools for users to fine-tune and customize the design plan according to their needs.

2. The clothing fashion trend prediction and intelligent design platform according to claim 1, characterized in that: The data acquisition module performs data acquisition through the following steps: Collect sales data from e-commerce platforms, including product sales volume, sales revenue, and best-selling products; Collecting review data from user review areas, including user evaluations, ratings, and review content of products; Collect discussion data from social media platforms, including topics, tags, post content, and user interaction data related to clothing; The press conference information is collected from the fashion press conference, and the press conference information includes the press conference time, brand, display clothing style, color, and material.

3. The clothing fashion trend prediction and intelligent design platform according to claim 2, characterized in that: The data collection module collects data regularly through web crawler technology and API interface, and stores the data in the database.

4. The clothing fashion trend prediction and intelligent design platform according to claim 1, characterized in that: The data processing module includes: The data cleaning unit is used to remove noise data, duplicate data, and incomplete data from the collected raw data, unify the data format, and fill in missing values; The variational mode decomposition unit is used to perform multi-scale decomposition on the cleaned data and extract various modal components.

5. The clothing fashion trend prediction and intelligent design platform according to claim 4, characterized in that: The formula of variational mode decomposition is: ； in, For the modal, is its center frequency.

6. The clothing fashion trend prediction and intelligent design platform according to claim 1, characterized in that: The gated recurrent unit model includes: Update gate, used to control the impact of the previous state on the current state; Reset gate, used to control the impact of current input on the current state; The current state unit is used to calculate the state at the current moment.

7. The clothing fashion trend prediction and intelligent design platform according to claim 6, characterized in that: The mathematical formula of the gated recurrent unit model is: ； ； ； ； in, To update the gate, To reset the gate, is the candidate state at the current moment, is the state at the current moment, W and b are model parameters.

8. The clothing fashion trend prediction and intelligent design platform according to claim 1, characterized in that: The mathematical formula for the generative adversarial network is: ； Among them, D is the discriminator, G is the generator, is the real data sample, z is the noise sample, is the distribution of real data, is the distribution of noise samples.

Images