Image enhancement for fringe projection profilometry

The fringe projection profilometry (FPP) is a popular approach for acquiring high-fidelity 3D model of objects. In the FPP, parallel fringe patterns are projected onto an object, and the deformed fringe as shown on the object surface is captured by a camera for 3D reconstruction. Due to the imperfection of imaging devices and operating environment, the captured fringe images however are inevitably to have many artifacts. They severely affect the robustness of the FPP and the quality of the reconstructed 3D model. In this paper, a new approach is proposed that successfully characterizes some common artifacts of fringe images, such as bias and noise, using the dual-tree complex wavelet transform. Effective methods are then devised to remove them from the fringe images. Experimental results show that the proposed algorithm is superior to the traditional methods and facilitates accurate reconstruction of objects' 3D model even with low quality fringe images.