基于多尺度卷积神经网络的非遗烙画设计方法及系统

By using a multi-scale convolutional neural network design method, combined with an intangible cultural heritage composition knowledge graph and an adaptive vision-process coupling network, a parameterized guidance scheme for intangible cultural heritage pyrography is generated. This solves the problem of the disconnect between design and process in traditional design methods, and realizes efficient and feasible intelligent design and human-machine collaborative optimization of intangible cultural heritage pyrography.

CN122415779APending Publication Date: 2026-07-17WUHAN INST OF TECH

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
WUHAN INST OF TECH
Filing Date
2026-03-09
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies lack a design scheme that can deeply integrate the aesthetic knowledge of intangible cultural heritage pyrography, intelligently generate visually realistic and technically feasible designs, and it is difficult to achieve human-computer collaborative interaction optimization. Traditional design methods suffer from problems such as a disconnect between design conception and technical implementation, difficulties in teaching and inheritance, and low innovation efficiency.

Method used

A design method for intangible cultural heritage pyrography based on multi-scale convolutional neural networks is adopted. By constructing a knowledge graph of intangible cultural heritage composition, using a multimodal large model to analyze creative input, integrating a few-sample attention mechanism to generate sketches, and generating a parameterized pyrography guidance scheme through an adaptive vision-process coupling network, human-computer collaboration is achieved by combining style fine-tuning and interaction optimization.

Benefits of technology

It achieves cultural authenticity, realistic texture, and technical feasibility in the design of intangible cultural heritage pyrography, lowers the learning threshold, improves design efficiency and innovation possibilities, supports low-latency interactive design, and promotes the digital inheritance and innovative development of intangible cultural heritage skills.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122415779A_ABST
    Figure CN122415779A_ABST
Patent Text Reader

Abstract

本发明涉及数字图像处理技术领域,具体涉及一种基于多尺度卷积神经网络的非遗烙画设计方法及系统。方法包括:获取用户自然语言创意与载体信息;利用融合非遗知识图谱的多模态大模型解析生成结构化构图描述;基于该描述,通过集成小样本注意力机制的扩散模型生成具有烙画笔触质感的初始设计草图;将草图与载体图像输入自适应视觉‑工艺耦合网络,该网络通过视觉特征通道与物理参数通道的深度融合,生成与载体及创作者风格相匹配的参数化烙制指导方案。本发明最终输出可直接指导烙制的详细方案,同时支持个性化风格学习与人机交互优化,有效提升了非遗烙画设计的智能化水平、创作效率与传承效果。
Need to check novelty before this filing date? Find Prior Art