Image enhancement and denoising for fringe projection patterns

Abstract

Fringe projection profilometry (FPP) is one of the widely studied for three dimensional (3D) surface reconstruction in academia because of the advantages of non-contact, full-field, fast and popularly used metrological tools, in which the imperfection of operating devices and environment can severely influence the images of the captured fringe patterns. To analyze and process low quality fringe patterns, noise reduction and fringe enhancement are key problems for fringe pattern processing. For acquiring accurate 3D surface reconstruction of objects, we proposed a new filter approach for noisy fringe patterns images, enhancement and processing of fringe patterns with parametric estimation is presented. Simulated and experimental results show that the proposed method can significantly facilitate image denoising and enhancement for fringe projection patterns, and then with the ability to obtain accurate 3D surface reconstruction of objects.