Simple construction of calibration curve and surface roughness detection in surface roughness detection by speckle contrast method

The Surface roughness is one of the important factors related to the appearance and performance of products, and the interest is further increased in various kinds of manufacturing. This paper describes a novel method for measuring surface roughness by laser speckle method. In the proposed method, instead of directly fitting with surface roughness and speckle feature amount, indirect parameters are obtained by converting the functional system. The polished metal pieces are used as the sample set. The prediction formula was obtained based on the reference value measured by the conventional method, and the prediction error of the prediction formula was verified by applying the sample of unknown surface roughness. In addition, we confirmed that the prediction accuracy improves by limiting the range for constructing the prediction formula. This method enables us to quickly detect the surface roughness by using a very simple configuration and a simple analysis method. A system that uses the method can be constructed calibration curve of surface roughness. © 2019 The Illuminating Engineering Institute of Japan. All rights reserved.