Abstract
Fringe projection profilometry (FPP) has been widely used in various fields because of its advantages of non-contact, non-destructive, high-precision, and high-speed. However, due to the subsurface scattering effect, FPP encounters severe challenges when performing three-dimensional (3D) measurements of translucent objects. Based on FPP, this paper proposes a novel 3D measurement method for translucent objects using periodic line-fringe patterns and complementary Gray code patterns. By photographing a set of periodic line-fringe patterns with different phases and performing multi-Gaussian sum fitting on the grayscale values of the same pixel position in different images, the wrapping phase of the fringe corresponding to the maximum fitted grayscale value can be obtained. Furthermore, a novel complementary Gray code method is proposed to solve the absolute phase, which avoids phase unwrapping errors caused by subsurface scattering. Comparative experimental results show that the point cloud of translucent objects obtained using the proposed 3D reconstruct method is more complete, denser, and smoother, and more accuracy. The fitting diameter errors of the reconstruction results of the translucent nylon sphere and PP sphere are 0.025 mm and 0.044 mm respectively. In addition, our method can use the conventional binocular stereo vision system based on grating projection, without modifying the hardware system.

This publication has 36 references indexed in Scilit: