3D Surface Roughness Measurement Using Shape From Focus

Surface roughness describes the micro-geometry shape feature of part surface, which is significative to the design, process and capability judgement. The traditional touched measurement methods are hurtful to the surface, and have problems of low mensuration coefficient. The traditional 2D surface roughness can’t objectively accurately describe the 3D shape feature. To this question, the paper adopts non-touched technology — shape from focus(SFF) to reconstruct the 3D surface shape, and measure 3D surface roughness. The measurement resulta compared with the international universal measurement equipment—Rtec.The research results of the paper are as following:(1) Designed a focus measurement function. In the SFF technology, radius of confusion circle effects the difference of neighbourhood pixels and resolution of images directly. The paper uses relative degree of grey incidence(RDGI) to measure focus level of pixel, proposed a focus mesurement function based RDGI(FMRDGI). The comparison experiment of reconstruction shape with Rtec shows, the proposed algorithm promotes on precision and time efficiency.(2) Proposed a shape from focus algorithm for memory efficiency. SFF has a connatural flow is large data, proposed spread algorithm for me mory efficiency on this question. The algorithm uses traditional method to compute the depth of spread region firstly, then search the accurate depth in the local range. The experience shows, the proposed algorithm can guar antee precision, and memory usage rate reduces 9.6% at the same time.(3) Designed and developed a 3D surface roughness measurement system. With plugins and configuration files development method, the system is expandable strongly. Measured 3D surface roughness parameters Sa and Sq, compared with measurement result of Rtec, it shows, our measurement system has precision of relative error less than 5%, can be used in practice application.