Mineral pigment recognition based on spectral feature analysis and quantification

Analysis and identification of the pigment composition of the cultural relics is the key and difficult point in the field of the protection of cultural relics. In this paper, we use hyperspectral technology to make identification and analysis of three kinds of Commonly used green mineral pigment composition unharmed The characteristic wavelength of the extreme point and the inflection point are found by calculating the first derivative and the second order derivative of the spectral curve. The experiment uses three common green mineral pigments of Chinese traditional painting to analysis the spectral absorption characteristics. The effective characteristics of these three kinds of pigments were calculated by correlation coefficient. Finally, based on the extracted features, the decision tree is established to distinguish the mineral pigments.