The storage duration of ancient glass cultural relics is too long storage,their chemical composition will inevitably change,and even the original appearance can not be identified,so it is of great significance to use scientific statistical methods to analyze and identify their composition.54 glass cultural relics are taken as the research objects.The chemical composition content,ornamentation and color of the glass cultural relics are selected.And the Chi-square independence test is used to explore the influencing factors of the surface weathering of glass cultural relics.It is found that the surface weathering of cultural relics has a strong correlation with the type of glass,but a weak correlation with the decoration and color.Then,in order to accurately judge the category of cultural relics,a method for identifying glass cultural relics based on Learning Vector Quantization(LVQ)network is proposed.Finally,the systematic clustering method is used to subclassify the two types of glasses,and the high-potassium glasses are divided into high-silicon high-potassium glasses,high-potassium high-potassium glasses and high-calcium high-potassium glasses.The lead barium glass is divided into high-silicon lead-barium glass and high-lead-type lead-barium glass.
维普
万方数据