基于规则学习的传统纹样统一生成模式研究

传统纹样的生成是图案生成领域最具代表性的研究方向之一，从传统纹样的构图规则出发生成纹样是最常用的方法之一。以传统纹样为研究对象，针对现有传统纹样生成方法泛化能力差的问题，通过对不同传统纹样进行特征分析，找出其生成规则的异同，提出了一种传统纹样统一生成模式：寓意嵌入的层次迭代生成模式。选取几何纹、动物纹、植物纹中的典型纹样，利用统一生成模式进行重构，验证了模式的有效性。实验结果表明，统一生成模式适合多种传统纹样的重构，一定程度上是对传统纹样重构的创新应用。

An unified generation scheme of traditional patterns based on rule learning

The generation of traditional patterns is one of the most representative research directions in the field of pattern generation.It is one of the most common methods to generate patterns based on the rules of composition of traditional patterns.To solve the problems of poor generalization ability of the existing pattern generation method,by conducting the feature analysis of different traditional patterns,the similarities and differences of the generation rules are found out,and the traditional pattern generation scheme of a hierarchical iterative generation pattern with embedded cultural implications is proposed .In order to verify the effectiveness of the algorithm,this paper selects typical patterns in geometric,animal and plant patterns for reconstruction using the generation scheme.Experimental results show that the proposed generation scheme can be applied to the reconstruction of a variety of different traditional patterns,and the results demonstrate an innovative application of traditional patterns to a certain extent.