Abstract: The Hyperspectral technology combined with stoichiometry was used to classify and identify the inkblots of 42 different brands and types of black dye pens commonly found on the market. Firstly,hyperspectral scanning of ink samples from 42 black dye pens in the range of 450-950 nm was performed. With the help of SPSS 26 software and Originpro2023,the systematic cluster analysis method was used to classify the ink samples,and the principal component analysis was used to verify the classification results. Systematic cluster analysis of hyperspectral data could divide 42 ink samples into two categories with a total of 7 categories,including 4 types of oil-based pens and 3 types of water-based pens. Hyperspectral technology combined with stoichiometry can classify and identify black dye pen ink,which provides technical support for the detection of black dye pen ink in actual cases.