Characterization Of Object Roughness Based On Mueller Matrix

In the field of industrial production,the detection of object surface roughness has always been a concern.Polarization detection technology is not limited by light intensity contrast,target background,surface structure and other factors,and it is non-contact detection,so it is also applied to the detection of material surface microstructure.Mueller matrix is a 4×4 matrix with 16 elements,which contains almost all microscopic information on the surface of an object and is a very important parameter in polarization detection technology.Therefore,this paper mainly studies the scattering Mueller matrix of random rough surface under different roughness,and uses Mueller matrix elements to characterize the roughness.The specific work is as follows:1.Gaussian random rough surface is generated by Monte Carlo method,and the scattering of light waves by random rough surface is analyzed by Kirchhoff approximation method.Combined with Stratton-Chu integral equation,the analytical expression of spatial scattering field is obtained.By analyzing the transformation relationship between Jones matrix and Mueller matrix,the 16 elements of Mueller matrix are related to scattering coefficient.2.The Mueller matrix of copper,nickel,iron,glass,gallium arsenide and lithium niobate with different roughness is numerically simulated,and the influence of the correlation length,root mean square height and relative roughness of rough surface on Mueller matrix elements is analyzed.The results show that the absolute values of Mueller matrix scattering in the incident plane and in the mirror direction of incident light are symmetrically distributed,with |m₀₀|=|m₁₁|,|m01|=|m10|,|m₂₂|=|m₃₃| and |m23|=|m32|.With the increase of roughness,the absolute values of all elements of Mueller matrix decrease,indicating that the rougher the material surface,the stronger the scattering,the smaller the scattered light intensity in the mirror direction of incident light,the stronger the depolarization ability,and the weaker the horizontal bidirectional attenuation ability and polarization ability.3.In Mueller matrix m₀₀?m₁₁?m₂₂?m₃₃ are sensitive to the roughness,and vary with roughness in a wide range.m₂₂ is selected as a parameter to characterize relative roughness,and the curve of m₂₂ varying with relative roughness is fitted by linear fitting method,nonlinear fitting method and Fourier fitting method respectively,and the expression to characterize relative roughness is obtained.Through comprehensive comparison,the nonlinear characterization method has high fitting degree,small error and easy calculation,and is suitable for characterization of material roughness.