For the concentration detection of particulate matter in the atmospheric environment, a Monte Carlo polarization scattering model was established on the basis of the Mie scattering theory, with the oil mist environment as the research object. The Stokes-Mueller matrix analysis method was employed to study the variation of Mueller matrix elements, and the patterns were investigated through Mueller matrix decomposition. Analyzing the Mueller matrix elements was an effective way of obtaining medium properties. Using the method based on oil mist concentration characterized by optical thickness, the model was proved correctly by experimental tests and simulations. The results indicate that with the increase in optical thickness (i.e., increase in concentration), the pattern of the Mueller matrix element M₁₁ enlarges, and the non-polarized optical characteristics intensify. Moreover, the intensity value of the scattering depolarization matrix decomposed by the Mueller matrix goes up, and the scattering depolarization ability is enhanced. This method can distinguish the characteristics of particulate matter media at different concentrations more intuitively and achieve a visualization effect. It provides a theoretical basis and a new test method for the detection of atmospheric particulate matter concentration.