Surface Roughness Measurement Based On Laser Triangulation

The surface roughness measurement is one important project in the mechanical manufacturing field. With the disadvantages of contact, scratching the surface measured easily, and so on, the filed to be applied of conventional stylus method is very limited. But, with the advantages of non-contact, wide adaptability of materials, the laser triangulation has a strong application potential. A surface roughness measurement method is studied in the paper, which is based on laser triangulation and digital image processing technique. In this measuring system, we use the line-structured light as light source, microscope lens and high-accuracy CCD sensor as displacement sensor as well.According to the theoretic deducting result of direct-type laser triangulation, relative factors affecting the measurement sensitivity are analyzed in this paper. The respective effect of measurement depth, optical magnification as well as working angle to the measurement sensitivity is qualitatively analyzed. Furthermore, the affecting relation between the working angle and the sensitivity is mainly analyzed, which plays a significant role in the system structure design. The results indicate that the work angle could not too small in view of the radial physical dimensions of imaging system and the laser; In addition, while the working angle is gradually increased to the bigger angle, it is needed to dynamically modify the working distance and focal length to achieve the definition imaging and hence leads to the optical magnification decrease. In this case, the combined influence of the optical magnification and the working angle to the sensitivity may be negative. To the direct-type measurement system in this paper, the corresponding working angle under the highest sensitivity equals 70° approximately.The extracting the centre-line of light-stripe image is the essential step to the method proposed in this paper. Based on the characters of line-structured light source and local feature of the measured surface, the laser stripe formed often meets some problems such as the uneven width and non-concentrated brightness and larger discreteness. This paper proposes an improved barycenter algorithm to extract the laser strip center line. Based on the adaptive binarization processing, the algorithm obtains all columns of normal widths and the light stripe’s accurate edge. Then, according to light stripe’s edge and gray data, the center of light stripe is calculated with the barycenter method. The result verified that the calculated range of barycenter in the improved method is consistent with light strip width, so it holds high extraction accuracy and good adaptability.The Gaussian filter is applied to the result of extracting the light stripe center-line, and then, the surface roughness profile curve is obtained. And, the surface roughness can be calculated, according to the evaluation principle of surface roughness.To verify the probability of this measure method, the nearly-standard planing surface is measured experimentally with the proposed method and the stylus method(Mitutoyo SJ-410) respectively. As for two methods, the profilogram of surface appearance are greatly similar in the shape and the amplitude, and also the roughness statistics values are close. This results verify that the laser triangulation based on the line-structured light can be applied to the surface roughness measurement. Moreover, this method has the advantages of rapid measurement velocity and the characteristic of directly indication surface roughness outline, and so it holds good development potential.