语义驱动的侗族织锦图案辅助设计路径研究

目的 优化民族图案创新设计过程中由于图案语义的隐性、复杂性等特点,造成创新设计的图案脱离原始语境,无法展现文化内涵.方法 以设计需求的关键性语义为驱动,分别利用TextRCNN、可拓语义量表、深度学习等技术手段,建立包括语义输入与转化、侗锦图案匹配、设计辅助3个层面的侗锦图案智能设计路径.结果 通过对设计需求关键性语义的提取,匹配相应语义的侗锦图案,并智能生成语义相关的设计方案.结论 以语义为驱动的民族图案辅助设计路径,有助于提高设计的效率,增强民族文化元素使用的准确性.

Aided Design Path of Dong Brocade Pattern Driven by Semantics

In the process of innovative design of national patterns, due to the implicit and complex features of pattern semantics, innovative design patterns are separated from the original context and unable to show the cultural connotation. The purpose of this paper is to optimize and improve the problem. Driven by the key semantics of design requirement, the intelligent design path of Dong brocade pattern including semantic input and transformation, Dong brocade pattern matching and design assistance is established by using TextRCNN, extension semantic scale, deep learning and other technical means. Through the extraction of the key semantics of design requirement, the Dong brocade patterns with corresponding semantics are matched, and the semantics-related design schemes are generated intelligently. The aided design path of ethnic patterns driven by semantics can improve the efficiency of design and increase the accuracy of the use of ethnic cultural elements.