深度学习视阈下湖湘印染艺术的模因溯源

Deep Learning-based Primal Memes Trace Back for Hunan Folk Printing and Dyeing

汉文化与少数民族文化交融是湖湘民间印染艺术的显著表征之一。通过采集中国历代传统纹样图像数据样本,基于迁移学习和分段训练策略构建深度学习神经网络分类模型,以民间代表性作品为对象,量化分析和探究湖湘印染艺术的中华文化模因传承机制。结合数据分析可推断,湖湘印染艺术的模因传承过程复杂,时间、功能性、文化自信、地域文化性格等是影响模因生命力的重要因素,而印染工艺对模因传承无实质性影响。以上研究成果有助于把握湖湘印染艺术的文化内涵和美学价值,促进该文化遗产在当代情境下的精准传承和创新利用。

The blending of Han culture and ethnic minority culture is one of the remarkable characteristics of Hunan folk printing and dyeing art.This paper focuses on the cultural meme tracing of Hunan printing and dyeing art via deep learning technologies.The research steps are as follows:Firstly,collecting the image data of traditional patterns in the past dynasties of China as samples;secondly,building a new convolution neural network classification model with the strategy of transfer learning and segmented training;thirdly,taking the representative works of Hunan printing and dyeing art as the research object;finally,quantitative analyzing the cultural meme inheritance code of Hunan printing and dyeing patterns.The research data reveals that the inheritance process of memes in Hunan printing and dyeing art is complex,and time,functionality,cultural self-confidence and regional cultural character are the important factors affecting the vitality of memes.Printing and dyeing process has no substantial impact on meme inheritance.The research results are helpful to grasp the cultural connotation and aesthetic value of Hunan printing and dyeing art,promoting the accurate inheritance and innovative utilization of the cultural heritage in the contemporary context.