Three-dimensional digital auxiliary system based on ceramic painting creation

Abstract

The invention discloses a three-dimensional digital auxiliary system based on ceramic painting creation. The three-dimensional digital auxiliary system comprises a three-dimensional scanning system used for converting a ceramic painting into a three-dimensional digital model; the digital image processing system is used for adjusting digital images of the ceramic painting works and repairing and enhancing the images; the three-dimensional modeling system is used for constructing a three-dimensional model of the ceramic painting works; the virtual reality system is used for displaying the three-dimensional model of the ceramic painting works; the artificial intelligence system is used for carrying out deep learning and analysis on the ceramic painting works and providing various auxiliary functions; according to the invention, a creator can quickly scan, process, model and virtually display the works through a digital technology, so that the creation period is shortened, the creation efficiency is improved, the system can intelligently evaluate and suggest the ceramic painting works, the creator is helped to discover and improve the defects, and the quality of the works is improved.

Description

Technical Field

[0001] The present invention relates to the field of ceramic painting creation auxiliary technology, and in particular to a three-dimensional digital auxiliary system based on ceramic painting creation. Background Art

[0002] Ceramic painting creation refers to applying painting art to the surface of ceramic objects, expressing the artist's aesthetic concepts and thoughts and feelings through various painting techniques and methods, and giving ceramic objects new artistic value and aesthetic connotations. The creation of ceramic painting needs to combine the characteristics of ceramic materials and use various painting tools, such as brushes, row brushes, brushes, etc., to create on the surface of ceramics. There are various techniques for ceramic painting, including overglaze color, underglaze color, glaze color and other methods, each of which has its unique characteristics and effects. In the creative process, artists need to have a deep understanding of the nature and characteristics of ceramic materials, master the skills and laws of various painting techniques, and also consider factors such as the shape, use and aesthetic needs of the objects, so as to create excellent works that meet the spirit of the times and aesthetic needs. Ceramic painting creation is not only a form of artistic expression, but also a precious carrier of human culture and history. Through ceramic painting creation, human history, culture and artistic achievements can be recorded and inherited, and ceramic art can also be innovated and developed to enrich people's spiritual and cultural life.

[0003] In the process of traditional ceramic painting creation, creators need to spend a lot of time and energy on manual production. In addition, creators may run out of inspiration during the creation process. Therefore, a three-dimensional digital auxiliary system based on ceramic painting creation is proposed. Summary of the invention

[0004] The purpose of the present invention is to solve the problems existing in the prior art that in the traditional ceramic painting creation process, the creator needs to spend a lot of time and energy on manual production. In addition, the creator may encounter the situation of running out of inspiration during the creation process. A three-dimensional digital auxiliary system based on ceramic painting creation is proposed.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] The three-dimensional digital auxiliary system based on ceramic painting creation includes:

[0007] 3D scanning system: used to convert ceramic paintings into 3D digital models;

[0008] Digital image processing system: used to adjust the digital images of ceramic paintings, as well as to repair and enhance the images;

[0009] 3D modeling system: used to construct 3D models of ceramic paintings;

[0010] Virtual reality system: used to display three-dimensional models of ceramic paintings;

[0011] Artificial intelligence system: used for deep learning and analysis of ceramic paintings, and provides various auxiliary functions;

[0012] The 3D scanning system converts the physical work into a high-precision 3D digital model, which is passed to the digital image processing system for further adjustment and optimization. The processed digital image and 3D model are input into the 3D modeling system to construct a 3D model of the ceramic painting. The 3D model of the ceramic painting is used for virtual reality display through the virtual reality system. The artificial intelligence system obtains data from the 3D scanning system, the digital image processing system, the 3D modeling system and the virtual reality system for analysis and learning. The artificial intelligence system identifies the style of the ceramic painting, generates creativity, and intelligently evaluates and suggests it, providing personalized support and suggestions for the creators. At the same time, the artificial intelligence system can also generate new creativity and design solutions for the creators to refer to and learn from.

[0013] The above technical solution further includes:

[0014] The three-dimensional scanning system includes a scanning device module, a control module, a data processing module, a three-dimensional reconstruction module and a data interaction module. The scanning device module includes a high-precision three-dimensional scanning device for scanning the surface of ceramic paintings and collecting geometric shapes and texture information of the works. The control module is used to manage and control the operation of the scanning device. The data processing module is used to process the raw data obtained from the scanning device. The three-dimensional reconstruction module converts the processed data into a three-dimensional digital model. The data interaction module is used to interact the generated three-dimensional model with other software or systems.

[0015] The control module communicates with the scanning device module by sending instructions and controls the operation and parameter setting of the scanning process. The scanning device module transmits the scanned data to the data processing module for subsequent processing. The data processing module transmits the processed data to the three-dimensional reconstruction module. The three-dimensional reconstruction module uses a reconstruction algorithm to convert the processed data into a three-dimensional digital model and extracts the geometry and texture information of the model.

[0016] Among them, the control module is responsible for communicating with the scanning device and sending control signals to control the speed and accuracy of the scan. The data processing module performs operations such as filtering, denoising, and registration on the data to improve data quality and accuracy. It uses a reconstruction algorithm to splice the scanned data into a complete three-dimensional model and extract the geometry and texture information of the model. The data interaction module provides a standard data format for data transmission and sharing with other software or systems.

[0017] The digital image processing system includes an input module, an image preprocessing module, an image segmentation module, a feature extraction module, an image restoration and enhancement module, an image synthesis module and an output module, wherein the input module is responsible for receiving digital image data of ceramic paintings, the image preprocessing module is responsible for preprocessing the input image, the image segmentation module is used to divide the image into different regions or objects, the feature extraction module is used to extract features from the image, the image restoration and enhancement module is used to restore and enhance the image, the image synthesis module is used to synthesize the restored and enhanced image, and the output module is used to save the processed image data into a file or output it in other ways for subsequent use or display;

[0018] Among them, preprocessing operations include image denoising, image enhancement, geometric correction, color correction, etc. to ensure image quality and accuracy. Image segmentation can be performed based on features such as color, texture, and edge to obtain detailed information on different parts of ceramic paintings. Features can be shape, texture, color or other statistical features for further image analysis and processing. Restoration operations can remove noise, scratches or other defects in images. Enhancement operations can increase the contrast, brightness, saturation, etc. of images to improve the visual effects of images. Synthesis can include recombining different image parts or fusing the repaired image with other images or backgrounds.

[0019] The three-dimensional modeling system includes an input module, a geometric modeling module, a material and texture module, an animation and deformation module, a lighting and rendering module and an output module, wherein the input module is responsible for receiving input data, the geometric modeling module is used to create and edit the geometric shape of the three-dimensional model, the material and texture module is used to specify the material and texture properties of the model, the animation and deformation module is used to add animation and deformation effects to the three-dimensional model, the lighting and rendering module is responsible for setting the lighting and rendering effects of the scene, the optimization module is responsible for optimizing the generated three-dimensional model to improve the accuracy and quality of the model, the optimization module transmits the optimized model to the output module for display and storage, and the output module is used to save the modeled three-dimensional model into a file or export it;

[0020] Among them, the input data can be scanned data, the creator's sketches, images or other forms of reference materials. The geometric modeling module provides basic geometric modeling tools, such as drawing lines, creating surfaces, Boolean operations, etc., to realize the geometric structure of the model. The material and texture module provides functions such as material library, texture generator and editor to achieve the appearance effect of the model. The animation and deformation module uses keyframe animation, skeletal animation or other technologies to achieve the dynamic performance of the model. The lighting and rendering module provides options such as lighting settings, ambient light, shadows and rendering parameters to achieve realistic presentation of the model. The optimization module optimizes the model's geometric details, texture mapping and lighting performance to meet the creator's needs and standards.

[0021] The virtual reality system includes an input module, a rendering module, an interaction module and an output module. The input module is responsible for receiving three-dimensional model data from other systems. The input module transmits the data to the rendering module for rendering and presentation. The rendering module is responsible for real-time rendering of the input three-dimensional model data to generate a three-dimensional scene in the virtual reality environment. The rendering module interacts with the input module, the interaction module and the output module, and transmits the rendering result to the output module for display. The interaction module is responsible for providing an interactive function between the user and the virtual reality environment. The output module is used to display the rendering result to the user.

[0022] Among them, the three-dimensional model data can be a digital model obtained by a scanning system, or a three-dimensional model constructed by a creator through a three-dimensional modeling system. The rendering module includes a graphics rendering engine and corresponding software, which can provide high-quality rendering effects and interactive functions.

[0023] The artificial intelligence system package and the interactive module generally include a data collection and processing module, a feature extraction and representation learning module, a style recognition module, a creative generation module, an intelligent evaluation and suggestion module and a personalized support and suggestion module, wherein the data collection and processing module is used to collect and organize relevant data of ceramic paintings, the feature extraction and representation learning module converts the image of the ceramic painting into a high-level abstract representation to capture the style, characteristics and visual information of the work, the style recognition module is used to identify the style or style elements of the ceramic painting, the creative generation module is used to generate ceramic paintings that conform to a specified style or theme, the creative generation module provides creative inspiration and auxiliary design for creators, the intelligent evaluation and suggestion module is used to evaluate and analyze ceramic paintings, and the personalized support and suggestion module is used to provide personalized support and suggestions according to the needs and preferences of the creators;

[0024] The data collection and processing module provides processed images and data to the feature extraction and representation learning module for training and learning. The feature extraction and representation learning module passes the extracted high-level abstract representation to the style recognition module for identifying and analyzing the style of the work. The creative generation module uses the results of the style recognition module to generate works that conform to the specified style. The intelligent evaluation and suggestion module uses the features extracted by the style recognition module and the user preference data provided by the personalized support and suggestion module to evaluate the work and provide suggestions. The personalized support and suggestion module provides personalized guidance and recommendations to other modules based on user needs and historical data.

[0025] Among them, relevant data include images, metadata and tags, etc. The data collection and processing module can also pre-process data to meet the needs of subsequent model training and analysis. The style or style elements include traditional, modern, abstract, etc. The style recognition module can help creators understand and analyze the style characteristics of the work, so as to provide relevant suggestions and references. The intelligent evaluation and suggestion module provides quality scores, technical guidance, improvement suggestions, etc. of the work to help creators improve the quality and creation level of the work. The personalized support and suggestion module provides creators with targeted creation guidance, material recommendations, etc. based on user input and historical data.

[0026] The feature extraction and representation learning module includes a data preprocessing submodule, a feature extraction algorithm submodule, a feature representation learning submodule, a feature selection and optimization submodule and a feature output submodule. The data preprocessing submodule is used to perform a series of preprocessing steps on the data. The feature extraction algorithm submodule is used to extract key features from ceramic paintings. The feature representation learning submodule is used to extract high-level abstract features from images. The feature selection and optimization submodule is used to select and optimize features related to the style and content of ceramic paintings during the feature extraction process. The feature output submodule: after extraction and selection optimization, the feature extraction module will output a set of key features, which can be used for subsequent style recognition and creative generation tasks;

[0027] The feature extraction algorithm submodule extracts key features from ceramic paintings:

[0028] Input preprocessing: First, perform necessary preprocessing operations on the input ceramic paintings to ensure the accuracy and consistency of the image data;

[0029] Convolutional layer: Use the convolutional layer to perform convolution operations on the image to extract local features in the image;

[0030] Pooling layer: The pooling layer is used to downsample the output of the convolutional layer to reduce the dimension of the feature and improve the robustness of the feature;

[0031] Fully connected layer: The fully connected layer integrates the features extracted by the previous convolutional layer and pooling layer to form a more abstract and advanced feature representation;

[0032] Back propagation and optimization: During the feature extraction process, the algorithm continuously updates the weight parameters through back propagation and optimization algorithms to minimize the prediction error and improve the accuracy of feature extraction. By training the deep neural network, the algorithm can automatically learn how to extract effective features from the original image;

[0033] Feature selection and fusion: After extracting a series of features, the algorithm needs to perform feature selection and fusion operations to select the features most relevant to the style and content of ceramic paintings;

[0034] Output features: After the above steps, the algorithm will output a set of key features that can reflect the style, theme and visual information of the ceramic painting;

[0035] The output features can be vectors, matrices, or higher-level semantic descriptions, depending on the needs of subsequent tasks. The preprocessing steps include denoising, enhancement, normalization, etc. to ensure the accuracy and consistency of the data. These key features include color, texture, shape, layout, etc. These high-level abstract features can capture the style, theme, and visual information of the work, providing a basis for subsequent style recognition and creative generation.

[0036] The steps of the creative generation module to generate ceramic paintings that conform to a specified style or theme are as follows:

[0037] Prepare a dataset: Prepare a suitable dataset that is large enough and representative enough to ensure that the generative model can learn the statistical laws of the data.

[0038] Build a generative model: Use a convolutional neural network or a recurrent neural network to build a model. The goal of the generative model is to generate realistic samples from random noise, so it is necessary to design a suitable network structure and loss function. The loss function usually uses the cross entropy loss function.

[0039] Initialize the generator and discriminator: Initialize the parameters of the generator and the discriminator. The input of the generator is random noise, and the output is a generated fake sample. The input of the discriminator is a real sample or a generated fake sample, and the output is the probability that the input sample comes from the real data set.

[0040] Training the generator and discriminator: By iteratively updating the parameters of the generator and discriminator, the discriminator can distinguish between real samples and fake samples, and the generator can deceive the discriminator so that the generated fake samples are as close to real samples as possible. During the training process, the generator is first fixed, the discriminator is trained, and then the discriminator is fixed and the generator is trained;

[0041] Evaluation and optimization: During the training process, the performance of the generator and discriminator is evaluated, and the hyperparameters and network structure are continuously adjusted. The evaluation indicators include but are not limited to the authenticity, diversity, and similarity of the generated samples with the real samples.

[0042] Application: Once trained, the GAN can be applied to a variety of tasks. By inputting different random noises or specific conditions, images or texts with similar styles or features can be generated.

[0043] The present invention has the following beneficial effects:

[0044] 1. In the present invention, through digital technology, creators can quickly scan, process, model and virtually display their works, thereby shortening the creation cycle and improving creation efficiency. The system can intelligently evaluate and make suggestions on ceramic paintings, helping creators to discover and improve deficiencies, thereby improving the quality of their works.

[0045] 2. In the present invention, through feature representation learning, the feature extraction module can automatically extract meaningful information from the original image and learn the style, theme and visual information of the work. These high-level abstract features can improve the accuracy and efficiency of subsequent tasks and provide better support and suggestions for creators. At the same time, through the deep learning generative model, the creative generation module can generate ceramic paintings that conform to a specific style or theme according to the needs and preferences of the creator. This automated creative generation technology provides creators with unlimited creative space and sources of inspiration, which helps to stimulate their creativity and improve their creative level. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a system block diagram of the three-dimensional digital auxiliary system based on ceramic painting creation proposed by the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] like Figure 1 As shown, the three-dimensional digital auxiliary system based on ceramic painting creation proposed by the present invention includes:

[0049] 3D scanning system: used to convert ceramic paintings into 3D digital models;

[0050] Digital image processing system: used to adjust the digital images of ceramic paintings, as well as to repair and enhance the images;

[0051] 3D modeling system: used to construct 3D models of ceramic paintings;

[0052] Virtual reality system: used to display three-dimensional models of ceramic paintings;

[0053] Artificial intelligence system: used for deep learning and analysis of ceramic paintings, and provides various auxiliary functions;

[0054] The 3D scanning system converts the physical work into a high-precision 3D digital model. The 3D digital model is passed to the digital image processing system for further adjustment and optimization. The processed digital image and 3D model are input into the 3D modeling system to construct a 3D model of the ceramic painting. The 3D model of the ceramic painting is used for virtual reality display through the virtual reality system. The artificial intelligence system obtains data from the 3D scanning system, digital image processing system, 3D modeling system and virtual reality system for analysis and learning. The artificial intelligence system recognizes the style of ceramic paintings, generates ideas, and intelligently evaluates and suggests ideas, providing creators with personalized support and suggestions. At the same time, the artificial intelligence system can also generate new ideas and design plans for creators to refer to and learn from.

[0055] The 3D scanning system includes a scanning device module, a control module, a data processing module, a 3D reconstruction module and a data interaction module. The scanning device module includes a high-precision 3D scanning device for scanning the surface of ceramic paintings and collecting geometric shapes and texture information of the works. The control module is used to manage and control the operation of the scanning device. The data processing module is used to process the raw data obtained from the scanning device. The 3D reconstruction module converts the processed data into a 3D digital model. The data interaction module is used to interact the generated 3D model with other software or systems.

[0056] The control module communicates with the scanning device module by sending instructions and controls the operation and parameter setting of the scanning process. The scanning device module transmits the scanned data to the data processing module for subsequent processing. The data processing module transmits the processed data to the 3D reconstruction module. The 3D reconstruction module uses the reconstruction algorithm to convert the processed data into a 3D digital model and extracts the geometry and texture information of the model.

[0057] Among them, the control module is responsible for communicating with the scanning device and sending control signals to control the speed and accuracy of the scan. The data processing module performs operations such as filtering, denoising, and registration on the data to improve data quality and accuracy. It uses a reconstruction algorithm to splice the scanned data into a complete three-dimensional model and extract the model's geometry and texture information. The data interaction module provides a standard data format for data transmission and sharing with other software or systems.

[0058] The digital image processing system includes an input module, an image preprocessing module, an image segmentation module, a feature extraction module, an image restoration and enhancement module, an image synthesis module and an output module. The input module is responsible for receiving digital image data of ceramic paintings. The image preprocessing module is responsible for preprocessing the input image. The image segmentation module is used to divide the image into different areas or objects. The feature extraction module is used to extract features from the image. The image restoration and enhancement module is used to restore and enhance the image. The image synthesis module is used to synthesize the restored and enhanced image. The output module is used to save the processed image data into a file or output it in other ways for subsequent use or display.

[0059] Among them, preprocessing operations include image denoising, image enhancement, geometric correction, color correction, etc. to ensure image quality and accuracy. Image segmentation can be performed based on features such as color, texture, and edge to obtain detailed information on different parts of ceramic paintings. Features can be shape, texture, color or other statistical features for further image analysis and processing. Restoration operations can remove noise, scratches or other defects in images. Enhancement operations can increase the contrast, brightness, saturation, etc. of images to improve the visual effects of images. Synthesis can include recombining different image parts or fusing the repaired image with other images or backgrounds.

[0060] The 3D modeling system includes an input module, a geometric modeling module, a material and texture module, an animation and deformation module, a lighting and rendering module, and an output module. The input module is responsible for receiving input data. The geometric modeling module is used to create and edit the geometric shape of the 3D model. The material and texture module is used to specify the material and texture properties of the model. The animation and deformation module is used to add animation and deformation effects to the 3D model. The lighting and rendering module is responsible for setting the lighting and rendering effects of the scene. The optimization module is responsible for optimizing the generated 3D model to improve the accuracy and quality of the model. The optimization module passes the optimized model to the output module for display and preservation. The output module is used to save the modeled 3D model into a file or export it.

[0061] Among them, the input data can be scanned data, the creator's sketches, images or other forms of reference materials. The geometric modeling module provides basic geometric modeling tools, such as drawing lines, creating surfaces, Boolean operations, etc., to realize the geometric structure of the model. The material and texture module provides material library, texture generator and editor and other functions to achieve the appearance effect of the model. The animation and deformation module uses keyframe animation, skeletal animation or other technologies to achieve the dynamic performance of the model. The lighting and rendering module provides options such as lighting settings, ambient light, shadows and rendering parameters to achieve realistic presentation of the model. The optimization module optimizes the model's geometric details, texture mapping and lighting performance to meet the creator's needs and standards.

[0062] The virtual reality system includes an input module, a rendering module, an interaction module and an output module. The input module is responsible for receiving 3D model data from other systems. The input module passes the data to the rendering module for rendering and presentation. The rendering module is responsible for real-time rendering of the input 3D model data to generate a 3D scene in the virtual reality environment. The rendering module interacts with the input module, the interaction module and the output module, and passes the rendering result to the output module for display. The interaction module is responsible for providing the interaction function between the user and the virtual reality environment. The output module is used to display the rendering result to the user.

[0063] Among them, the three-dimensional model data can be a digital model obtained by a scanning system, or a three-dimensional model constructed by a creator through a three-dimensional modeling system. The rendering module includes a graphics rendering engine and corresponding software, which can provide high-quality rendering effects and interactive functions.

[0064] The interactive module of the artificial intelligence system package generally includes a data collection and processing module, a feature extraction and representation learning module, a style recognition module, a creative generation module, an intelligent evaluation and suggestion module and a personalized support and suggestion module. The data collection and processing module is used to collect and organize relevant data of ceramic paintings. The feature extraction and representation learning module converts the image of the ceramic painting into a high-level abstract representation to capture the style, characteristics and visual information of the work. The style recognition module is used to identify the style or style elements of the ceramic painting. The creative generation module is used to generate ceramic paintings that conform to the specified style or theme. The creative generation module provides creators with creative inspiration and auxiliary design. The intelligent evaluation and suggestion module is used to evaluate and analyze ceramic paintings. The personalized support and suggestion module is used to provide personalized support and suggestions based on the needs and preferences of the creators.

[0065] The data collection and processing module provides processed images and data to the feature extraction and representation learning module for training and learning. The feature extraction and representation learning module passes the extracted high-level abstract representation to the style recognition module for identifying and analyzing the style of the work. The creative generation module uses the results of the style recognition module to generate works that conform to the specified style. The intelligent evaluation and suggestion module uses the features extracted by the style recognition module and the user preference data provided by the personalized support and suggestion module to evaluate and provide suggestions for the works. The personalized support and suggestion module provides personalized guidance and recommendations to other modules based on user needs and historical data.

[0066] Among them, relevant data include images, metadata and tags, etc. The data collection and processing module can also pre-process data to meet the needs of subsequent model training and analysis. The style or style elements include traditional, modern, abstract, etc. The style recognition module can help creators understand and analyze the style characteristics of the work, so as to provide relevant suggestions and references. The intelligent evaluation and suggestion module provides quality ratings, technical guidance, improvement suggestions, etc. of the work to help creators improve the quality and creative level of the work. The personalized support and suggestion module provides creators with targeted creative guidance, material recommendations, etc. based on user input and historical data.

[0067] The feature extraction and representation learning module includes a data preprocessing submodule, a feature extraction algorithm submodule, a feature representation learning submodule, a feature selection and optimization submodule, and a feature output submodule. The data preprocessing submodule is used to perform a series of preprocessing steps on the data. The feature extraction algorithm submodule is used to extract key features from ceramic paintings. The feature representation learning submodule is used to extract high-level abstract features from images. The feature selection and optimization submodule is used to select and optimize features related to the style and content of ceramic paintings during the feature extraction process. The feature output submodule: After extraction and selection optimization, the feature extraction module will output a set of key features, which can be used for subsequent style recognition and creative generation tasks;

[0068] The steps of the feature extraction algorithm submodule to extract key features from ceramic paintings are as follows:

[0069] Input preprocessing: First, perform necessary preprocessing operations on the input ceramic paintings to ensure the accuracy and consistency of the image data;

[0070] Convolutional layer: Use the convolutional layer to perform convolution operations on the image to extract local features in the image;

[0071] Pooling layer: The pooling layer is used to downsample the output of the convolutional layer to reduce the dimension of the feature and improve the robustness of the feature;

[0072] Fully connected layer: The fully connected layer integrates the features extracted by the previous convolutional layer and pooling layer to form a more abstract and advanced feature representation;

[0073] Back propagation and optimization: During the feature extraction process, the algorithm continuously updates the weight parameters through back propagation and optimization algorithms to minimize the prediction error and improve the accuracy of feature extraction. By training the deep neural network, the algorithm can automatically learn how to extract effective features from the original image;

[0074] Feature selection and fusion: After extracting a series of features, the algorithm needs to perform feature selection and fusion operations to select the features most relevant to the style and content of ceramic paintings;

[0075] Output features: After the above steps, the algorithm will output a set of key features that can reflect the style, theme and visual information of the ceramic painting;

[0076] The output features can be vectors, matrices, or higher-level semantic descriptions, depending on the needs of subsequent tasks. The preprocessing steps include denoising, enhancement, normalization, etc. to ensure the accuracy and consistency of the data. These key features include color, texture, shape, layout, etc. These high-level abstract features can capture the style, theme, and visual information of the work, providing a basis for subsequent style recognition and creative generation.

[0077] The steps for the creative generation module to generate ceramic paintings that meet the specified style or theme are as follows:

[0078] Prepare a dataset: Prepare a suitable dataset that is large enough and representative enough to ensure that the generative model can learn the statistical laws of the data.

[0079] Build a generative model: Use a convolutional neural network or a recurrent neural network to build a model. The goal of the generative model is to generate realistic samples from random noise, so it is necessary to design a suitable network structure and loss function. The loss function usually uses the cross entropy loss function.

[0080] Initialize the generator and discriminator: Initialize the parameters of the generator and discriminator. The input of the generator is random noise, and the output is the generated fake sample. The input of the discriminator is a real sample or a generated fake sample, and the output is the probability that the input sample comes from the real data set.

[0081] Training the generator and discriminator: By iteratively updating the parameters of the generator and discriminator, the discriminator can distinguish between real samples and fake samples, and the generator can deceive the discriminator so that the generated fake samples are as close to real samples as possible. During the training process, the generator is first fixed, the discriminator is trained, and then the discriminator is fixed and the generator is trained;

[0082] Evaluation and optimization: During the training process, the performance of the generator and discriminator is evaluated, and the hyperparameters and network structure are continuously adjusted. The evaluation indicators include but are not limited to the authenticity, diversity, and similarity of the generated samples with the real samples.

[0083] Application: Once trained, the GAN can be applied to a variety of tasks. By inputting different random noises or specific conditions, images or texts with similar styles or features can be generated.

[0084] In an embodiment of the present invention, a ceramic painting is placed on a scanning platform, a control module controls a scanning device to perform a three-dimensional scan of the work, a scanning device module converts the surface morphology and detail data of the work into point cloud or mesh data, and stores it as a digital model, a data processing module is used to clean and repair the digital model to remove noise and errors that may exist during the scanning process, and a three-dimensional reconstruction module is used to convert the processed data into a three-dimensional digital model;

[0085] Import the digital image of the ceramic painting into the digital image processing system, adjust the color, brightness, contrast and other aspects of the image to obtain a more accurate or artistic image performance. Use the image restoration and enhancement module to perform local restoration as needed to remove defects or damage in the image and enhance the details or texture of the image to make the work clearer, richer or more eye-catching. The final image is exported to various formats through the output module for easy integration or display with other systems.

[0086] Create an initial model in the 3D modeling software based on the scan data of the 3D scanning system or the creator's sketch. Use the modeling tools and functions provided by the software to adjust, refine and modify the model to accurately simulate the form and structure of the ceramic painting. Add details and textures through the material and texture module, add animation and deformation effects through the animation and deformation module, and set the lighting and rendering effects of the scene through the lighting and rendering module to make the model more realistic and detailed. After the model is completed, export it to a standard 3D model format, such as OBJ, FBX, etc., through the output module to facilitate integration or display with other systems;

[0087] Switch the 3D model of the ceramic painting to the virtual reality platform. Users can observe, rotate, zoom, move and modify it through the functions and interactive means in the virtual reality system to get a more comprehensive understanding of the characteristics and details of the work.

[0088] Using the algorithm library and machine learning framework, deep learning and analysis are performed on the data of ceramic paintings. By training the model, the artificial intelligence system can identify the specific style of ceramic paintings and classify them. According to the style characteristics and related data of ceramic paintings, the artificial intelligence system can generate creativity, provide creative inspiration and suggestions. For the specific requirements or goals of ceramic paintings, the artificial intelligence system can perform intelligent evaluation and suggestions to help creators improve the quality and artistic value of their works. The artificial intelligence system can be integrated with other systems, such as 3D scanning, digital image processing, and 3D modeling systems, to provide more comprehensive and personalized auxiliary functions and services.

[0089] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A three-dimensional digital auxiliary system based on ceramic painting creation, characterized in that: include: 3D scanning system: used to convert ceramic paintings into 3D digital models; Digital image processing system: used to adjust the digital images of ceramic paintings, as well as to repair and enhance the images; 3D modeling system: used to construct 3D models of ceramic paintings; Virtual reality system: used to display three-dimensional models of ceramic paintings; Artificial intelligence system: used for deep learning and analysis of ceramic paintings, and provides various auxiliary functions; The 3D scanning system converts the physical work into a high-precision 3D digital model, which is passed to the digital image processing system for further adjustment and optimization. The processed digital image and 3D model are input into the 3D modeling system to construct a 3D model of the ceramic painting. The 3D model of the ceramic painting is used for virtual reality display through the virtual reality system. The artificial intelligence system obtains data from the 3D scanning system, the digital image processing system, the 3D modeling system and the virtual reality system for analysis and learning. The artificial intelligence system identifies the style of the ceramic painting, generates creativity, and intelligently evaluates and suggests it, providing personalized support and suggestions for the creators. At the same time, the artificial intelligence system can also generate new creativity and design solutions for the creators to refer to and learn from.

2. The three-dimensional digital auxiliary system based on ceramic painting creation according to claim 1 is characterized in that: The three-dimensional scanning system includes a scanning device module, a control module, a data processing module, a three-dimensional reconstruction module and a data interaction module. The scanning device module includes a high-precision three-dimensional scanning device for scanning the surface of ceramic paintings and collecting geometric shapes and texture information of the works. The control module is used to manage and control the operation of the scanning device. The data processing module is used to process the raw data obtained from the scanning device. The three-dimensional reconstruction module converts the processed data into a three-dimensional digital model. The data interaction module is used to interact the generated three-dimensional model with other software or systems. The control module communicates with the scanning device module by sending instructions and controls the operation and parameter setting of the scanning process. The scanning device module transmits the scanned data to the data processing module for subsequent processing. The data processing module transmits the processed data to the three-dimensional reconstruction module. The three-dimensional reconstruction module uses a reconstruction algorithm to convert the processed data into a three-dimensional digital model and extracts the geometry and texture information of the model.

3. The three-dimensional digital auxiliary system based on ceramic painting creation according to claim 1 is characterized in that: The digital image processing system includes an input module, an image preprocessing module, an image segmentation module, a feature extraction module, an image restoration and enhancement module, an image synthesis module and an output module. The input module is responsible for receiving digital image data of ceramic paintings, the image preprocessing module is responsible for preprocessing the input image, the image segmentation module is used to divide the image into different areas or objects, the feature extraction module is used to extract features from the image, the image restoration and enhancement module is used to restore and enhance the image, the image synthesis module is used to synthesize the restored and enhanced image, and the output module is used to save the processed image data into a file or output it in other ways for subsequent use or display.

4. The three-dimensional digital auxiliary system based on ceramic painting creation according to claim 1 is characterized in that: The three-dimensional modeling system includes an input module, a geometric modeling module, a material and texture module, an animation and deformation module, a lighting and rendering module and an output module. The input module is responsible for receiving input data, the geometric modeling module is used to create and edit the geometric shape of the three-dimensional model, the material and texture module is used to specify the material and texture properties of the model, the animation and deformation module is used to add animation and deformation effects to the three-dimensional model, the lighting and rendering module is responsible for setting the lighting and rendering effects of the scene, the optimization module is responsible for optimizing the generated three-dimensional model to improve the accuracy and quality of the model, the optimization module passes the optimized model to the output module for display and storage, and the output module is used to save the modeled three-dimensional model into a file or export it.

5. The three-dimensional digital auxiliary system based on ceramic painting creation according to claim 1 is characterized in that: The virtual reality system includes an input module, a rendering module, an interaction module and an output module. The input module is responsible for receiving three-dimensional model data from other systems. The input module passes the data to the rendering module for rendering and presentation. The rendering module is responsible for real-time rendering of the input three-dimensional model data to generate a three-dimensional scene in a virtual reality environment. The rendering module interacts with the input module, the interaction module and the output module, and passes the rendering result to the output module for display. The interaction module is responsible for providing an interactive function between a user and the virtual reality environment, and the output module is used to display the rendering result to the user.

6. The three-dimensional digital auxiliary system based on ceramic painting creation according to claim 1 is characterized in that: The artificial intelligence system includes a data collection and processing module, a feature extraction and representation learning module, a style recognition module, a creative generation module, an intelligent evaluation and suggestion module, and a personalized support and suggestion module. The data collection and processing module is used to collect and organize relevant data of ceramic paintings. The feature extraction and representation learning module converts the image of the ceramic painting into a high-level abstract representation to capture the style, characteristics and visual information of the work. The style recognition module is used to identify the style or style elements of the ceramic painting. The creative generation module is used to generate ceramic paintings that conform to a specified style or theme. The creative generation module provides creators with creative inspiration and auxiliary design. The intelligent evaluation and suggestion module is used to evaluate and analyze ceramic paintings. The personalized support and suggestion module is used to provide personalized support and suggestions based on the needs and preferences of the creators. The data collection and processing module provides processed images and data to the feature extraction and representation learning module for training and learning. The feature extraction and representation learning module passes the extracted high-level abstract representation to the style recognition module for identifying and analyzing the style of the work. The creative generation module uses the results of the style recognition module to generate works that conform to the specified style. The intelligent evaluation and suggestion module uses the features extracted by the style recognition module and the user preference data provided by the personalized support and suggestion module to evaluate the works and provide suggestions. The personalized support and suggestion module provides personalized guidance and recommendations to other modules based on user needs and historical data.

7. The three-dimensional digital auxiliary system based on ceramic painting creation according to claim 6 is characterized in that: The feature extraction and representation learning module includes a data preprocessing submodule, a feature extraction algorithm submodule, a feature representation learning submodule, a feature selection and optimization submodule and a feature output submodule. The data preprocessing submodule is used to perform a series of preprocessing steps on the data. The feature extraction algorithm submodule is used to extract key features from ceramic paintings. The feature representation learning submodule is used to extract high-level abstract features from images. The feature selection and optimization submodule is used to select and optimize features related to the style and content of ceramic paintings during the feature extraction process. The feature output submodule: after extraction and selection optimization, the feature extraction module will output a set of key features, which can be used for subsequent style recognition and creative generation tasks; The feature extraction algorithm submodule extracts key features from ceramic paintings: Input preprocessing: First, perform necessary preprocessing operations on the input ceramic paintings to ensure the accuracy and consistency of the image data; Convolutional layer: Use the convolutional layer to perform convolution operations on the image to extract local features in the image; Pooling layer: The pooling layer is used to downsample the output of the convolutional layer to reduce the dimension of the feature and improve the robustness of the feature; Fully connected layer: The fully connected layer integrates the features extracted by the previous convolutional layer and pooling layer to form a more abstract and advanced feature representation; Back propagation and optimization: During the feature extraction process, the algorithm continuously updates the weight parameters through back propagation and optimization algorithms to minimize the prediction error and improve the accuracy of feature extraction. By training the deep neural network, the algorithm can automatically learn how to extract effective features from the original image; Feature selection and fusion: After extracting a series of features, the algorithm needs to perform feature selection and fusion operations to select the features most relevant to the style and content of ceramic paintings; Output features: After the above steps, the algorithm will output a set of key features that can reflect the style, theme and visual information of the ceramic painting.

8. The three-dimensional digital auxiliary system based on ceramic painting creation according to claim 6 is characterized in that: The steps of the creative generation module to generate ceramic paintings that conform to a specified style or theme are as follows: Prepare a dataset: Prepare a suitable dataset; the dataset should be large enough and representative enough to ensure that the generative model can learn the statistical laws of the data. Build a generative model: Use a convolutional neural network or a recurrent neural network to build a model; Initialize the generator and discriminator: Initialize the parameters of the generator and the discriminator. The input of the generator is random noise, and the output is a generated fake sample. The input of the discriminator is a real sample or a generated fake sample, and the output is the probability that the input sample comes from the real data set. Train the generator and discriminator: By iteratively updating the parameters of the generator and discriminator, the discriminator can distinguish between real samples and fake samples, while the generator can deceive the discriminator so that the generated fake samples are as close to real samples as possible; Evaluation and optimization: During the training process, the performance of the generator and discriminator is evaluated, and the hyperparameters and network structure are continuously adjusted; Application: Once trained, the GAN can be applied to a variety of tasks. By inputting different random noises or specific conditions, images or texts with similar styles or features can be generated.

Images