Lossless and comprehensive identification method based on visible spectrum imaging and used for ancient fresco pigment

Abstract

The invention relates to a lossless and comprehensive identification method based on visible spectrum imaging and used for ancient fresco pigment. The method includes: building a comprehensive frescopigment database according to basic pigment used by a fresco and a double-constant Kubelka-Munk theory to serve as the reference foundation of fresco pigment identification; acquiring the visible spectrum image of an ancient fresco; rendering the visible spectrum image to obtain a corresponding colored RGB image; using a superpixel segmentation algorithm to perform fine segmentation on the coloredRGB image; calculating the average spectral reflectance of each subregion; extracting the spectrum features of database samples and subregion pigment; building a fresco pigment identification methodto identify substance variety of the pigment of each subregion, further identifying the average particle size of the pigment if the pigment is single-color pigment, judging that the pigment is the pigment with minimum particle size if the pigment is mixed pigment, and further calculating the concentration proportion of each pigment; acquiring the two-dimensional distribution diagram of each pigment in the fresco so as to complete the lossless and comprehensive identification, which includes pigment substances, particle size, concentration and distribution, of the ancient fresco pigment.

Description

technical field [0001] The invention belongs to the technical field of historical and cultural heritage protection, and in particular relates to a non-destructive and comprehensive identification method for ancient mural paints based on visible spectrum imaging. Background technique [0002] my country has a large number of ancient painted murals, such as Dunhuang Mogao Grottoes, Shaanxi Yulin and Xinjiang Kizil murals, etc. They are an important part of ancient Chinese civilization. These ancient murals contain ancient Chinese politics, economy, culture, science and technology and Important historical information such as the development of production technology carries splendid civilization and inherits historical culture. It is a precious resource for studying ancient humanities, sociology, aesthetics and natural science in my country. However, due to the multiple influences of the natural environment and human factors, most of the mural paint layers in various parts of my ...

AI Technical Summary

Problems solved by technology

[0004] In terms of ancient mural pigment identification, traditional pigment identification is mostly based on lossy sampling analysis methods, such as X-ray diffraction and X-ray fluorescence analyzers, scanning electron microscopes, energy chromatographs, polarizing microscopes, Fourier transform infrared Instruments and equipment such as spectrometers, Raman spectrometers, Raman microspectrometers, laser-induced breakdown spectrometers, and optical microscopes analyze the material composition information of sampled materials. Such sampling and analysis methods cause irreversible damage to precious historical relics, and It can only identify the material composition of a single point, and cannot conduct a comprehensive analysis of the pigment distribution of the mural and the physical properties of the pigment, which has great limitations

In recent years, in-situ non-destructive analysis technology has been paid more and more attention by cultural relics protection workers, and equipment such as fiber optic reflectance spectrometer and portable Raman fiber spectrometer have been produced, which makes the analysis of pigments without destructive sampling. The risk of damage to cultural relics, but the instrument still needs to be in contact with the cultural relics during use, and can only perform single-point analysis, which still has certain limitations in the analysis and application of mural paints

In addition, the current pigment analysis mainly focuses on the identification of pigment material components, while ignoring factors such as the particle size of the pigment that affects the color of the mural and the concentration of each basic pigment in the mixed pigment, and the analysis of the mural pigment is not comprehensive enough.

[0005] For the above limitations in the identification of ancient mural pigments, neither the cultural and cultural circles nor the academic circles have proposed a reasonable and effective solution.

Images