Accurate color matching of mineral pigments is a key to attaining high-quality restoration and high-fidelity reproduction of the cultural heritage murals. The particle size of mineral pigments is an important factor affecting color information and spectral reflectance. Accurately obtaining the spectral reflectance of mineral pigments on the mural surface facilitates the identification of pigment particle size. However, spectral information which is disturbed cannot match with the spectral database of mineral pigment particle sizes accurately. Therefore, determining the effective particle size from the spectral information is impossible. The ratio derivative method is proposed to compensate for the disturbance. The spectral information is converted from spectral reflectance space to ratio derivative spectrum space, which reduces disturbances and enhances the spectra information characteristics of mineral pigments. Spectral matching is then performed in the ratio derivative spectrum space. In the experiments, Azurite and Malachite mineral pigments, which are frequently used in murals, were utilized as color samples disturbed by substrate and white pigments. The proposed method was used to analyze the experimental sample data. Results of spectral angle measurement and matching of the spectral curve in the ratio derivative space showed that matching accuracy can meet the requirements, verifying the validity of the method. The proposed method solves the problem of inaccurate matching of the disturbed spectral reflectance and provides accurate particle size reference information for color matching of mineral pigments during mural restoration.