3D shape measurement for shiny surface using pixel-wise composed fringe pattern based on multi-intensity matrix projection of neighborhood pixels

Abstract

Phase shifting profilometry (PSP) has been commonly used in three-dimensional shape measurement. However, image saturation is a challenging problem leading to phase and measurement errors when measuring the shiny surface. We propose a new method using pixel-wise composed fringe pattern based on multi-intensity matrix projection of neighborhood pixels. First, a multi-intensity projection matrix is established and projected onto the object. The optimal intensity (the maximum projection intensity without saturation) for each pixel and the mask of each intensity can be obtained by analyzing the proportion of saturated pixels in the captured matrix region. Then, a set of new fringe pattern images without saturation can be obtained with phase shifted sinusoidal fringe patterns of different intensity levels and the corresponding mask. With PSP, the object can be reconstructed. Compared with existing methods, the proposed method can significantly reduce the number of projection operations and time consumption. The performance of the proposed method is verified by the experiments.