摘要
目的以人工智能技术为工具,挖掘中国古代家具形态演变及发展脉络,借助智能技术完成家具形态谱系的整理、提取与归纳,解决传统人工收集分类效率低下问题。方法以宋代以来的椅类家具为例,首先利用机器视觉技术完成宋元时期,明代时期和清代时期的椅类家具形态特征度量分析;然后综合特征度量结果,应用人工智能图像分类技术开展古代家具图像分类工作;最后引入知识图谱与数据库的技术,构建古代家具形态知识网络,建立起中国古代家具形态谱系数据库。结论针对宋元、明代、清代时期的椅类家具进行特征提取与分类,并建立相对应的家具形态谱系库,多维度呈现古代家具形态嬗变历程,为我国古代家具的研究、传承与数字化保护提供参考,拓展人工智能在传统文化领域研究的应用。
With artificial intelligence technology as a tool,the work aims to dig out the evolution and development of an-cient Chinese furniture,and sort out,extract and summarize the morphological genealogy of furniture by artificial intelligence technology,to solve the problem of low efficiency of traditional manual collection and classification.Firstly,with the chair fur-niture from the Song Dynasty as an example,the machine vision technology was used to complete the morphological feature measurement and analysis of chair furniture in the Song,Yuan,Ming and Qing Dynasties.Then,the feature measurement results were integrated and artificial intelligence image classification technology was applied to carry out image classification of ancient furniture.Finally,the knowledge graphs and databases were introduced to construct the knowledge network of ancient furniture forms and establish the genealogy database of ancient Chinese furniture forms.Feature extraction and classification of chair fur-niture in the Song,Yuan,Ming,and Qing Dynasties,establishment of a corresponding furniture form genealogy database and multi-dimensional presentation of the evolution of ancient furniture forms provide a basis for the research,inheritance and digi-tal protection of ancient furniture in China and expand the application of artificial intelligence in the field of traditional culture research.
作者
苏晨
刘品卓
郑佳勇
郑晓如
SU Chen;LIU Pinzhuo;ZHENG Jiayong;ZHENG Xiaoru(Hubei University of Technology,Wuhan 430068,China)
出处
《包装工程》
CAS
北大核心
2024年第22期183-192,共10页
Packaging Engineering
基金
2021年湖北省教育厅项目——哲学社会科学研究重点项目(ZXKY202155)。
关键词
人工智能
中国古代家具
形态谱系
数据库
知识图谱
artificial intelligence
ancient Chinese furniture
morphological genealogy
database
knowledge graph