摘要
出土玻璃制品是丝绸之路中古代中西文化交流的重要物证.玻璃制品易受埋藏环境的影响而风化,导致其原有成分比例发生变化,从而影响对它类别的正确判断.为了识别古玻璃的化学成分和物理特征,采用统计学方法对文物表面有无风化与颜色、纹饰和类型的相关性和14种化学成分间的相关性进行分析,运用Logit模型融合极大似然估计研究表面化学成分含量统计规律,对风化前的化学成分含量进行预测,预测准确度达到84.6%,并引入随机森林分类模型帮助判断未知古代玻璃的分类.实验结果表明,该方法能有效鉴别出古代玻璃制品的分类.
Glassware is an important material evidence of ancient Chinese and Western cultural exchanges on the Silk Road.Ancient glass is easily weathered by the influence of buried environment,which leads to the change of its composition ratio,thus affecting the correct judgment of its category.In order to identify the chemical composition and physical characteristics of ancient glass,the statistical methods were used to analyze the correlation between weathering on the surface of cultural relics and color,ornamentation and type,as well as the correlation between 14 chemical components.Logit model combined with maximum likelihood estimation was used to study the statistical law of surface chemical composition content,and the chemical composition content before weathering was predicted.The prediction accuracy reached 84.6%.The random forest classification model was introduced to help judge the classification of unknown ancient glass.Experimental results show that this method can effectively identify the classification of ancient glass products.
作者
杜世强
乐靖雯
周海丽
杨松杰
DU Shi-qiang;YUE Jing-wen;ZHOU Hai-li;YANG Song-jie(College of Mathematics and Computer Science,Northwest Minzu University,Lanzhou 730030,China)
出处
《西北民族大学学报（自然科学版）》
2023年第2期8-16,共9页
Journal of Northwest Minzu University(Natural Science)
基金
2021年西北民族大学教育教学改革研究一般项目(2021XJYBJG-69)。
