摘要
民族文化图案是中华民族宝贵财富,对民族文化图案进行语义标注与分析,是挖掘其文化价值,进行再创作与应用的基础.本文以此为研究对象,在多分类字典学习的基础上,提出了一种多标签字典学习标注算法SCMIDL.算法结合字典不相关性与系数相似性,有效提高了多标签标注性能,实现了民族文化图案的自动标注.在收集并构建的三类民族文化图案数据集进行多标签语义标注实验,实验结果验证了算法的有效性.
The national cultural pattern is a precious treasure of the Chinese nation. Semantic annotation and analysis of national cultural pattern is useful in cultural heritage and modern recreation. This article presents a multi-label dictionary learning algorithm called similar coefficient multi-label incoherent dictionary learning(SCMIDL) based on a multi-class dictionary learning algorithm. SCMIDL combines the incoherence of the dictionary and the similarity of coefficients, which significantly improves the performance of multi-label annotation.The superior annotation ability of the algorithm was confirmed on three kinds of national cultural pattern datasets constructed for evaluation purposes.
作者
赵海英
陈洪
贾耕云
郑桥
王绍杰
Haiying ZHAO;Hong CHEN;Gengyun JIA;Qiao ZHENG;Shaojie WANG(Mobile Media and Cultural Computing Key Laboratory,Beijing Century College,Beijing University of Posts and Telecommunications,Beijing 102101,China;Institute of Network Technology,Beijing University of Posts and Telecommunications,Beijing 100876,China)
出处
《中国科学：信息科学》
CSCD
北大核心
2019年第2期172-187,共16页
Scientia Sinica(Informationis)
基金
北京市科委基金(批准号:D171100003717003)资助项目
关键词
民族文化图案
字典学习
多标签
图像标注
相关性
national cultural pattern
dictionary learning
multi label
image annotation
correlation
