High-reflective surfaces shape measurement technology based on adaptive fringe projection

Abstract

Fringe projection profilometry (FPP) is limited by the pixel depth of the camera and cannot correctly reconstruct complex objects with highly reflective surfaces. In this paper, we proposed an improved adaptive fringe technique to avoid image saturation. This method only needs to project a uniform gray level image and a set of horizontal and vertical phase-shifting fringe patterns to generate adaptive fringes. The gray level of saturated regions can be fitted by the uniform image and the coordinates of the areas are obtained by calculating the horizontal and vertical fringes. To avoid the process of recursion, we introduce the concept of the grayscale correction function. Furthermore, a modified mapping formula is adopted and refined, which enhances the mapping effect and reconstruction accuracy. Compared with the original adaptive fringe projection (AFP), the modified method further decreases the number of projected fringe patterns. We analyzed the error between the fitted value and the standard value to evaluate the performance of the method. Experimental results demonstrated that the proposed method maintains the accuracy of reconstruction while improving time efficiency compared to other high dynamic range (HDR) measurement methods. • The contribution of different fringe projection technology with highly reflective objects is investigated. • A novel adaptive fringe is proposed to measure the depth information of HDR objects. • An idea is proposed for the improvement of adaptive fringe projection.