Inpainting For Fringe Projection Profilometry Based on Iterative Regularization

This paper presents a new inpainting algorithm for fringe projection profilometry (FPP) based on the iterative regularization technique. Conventional FPP approach often cannot accurately reconstruct the 3D model of objects in the presence of highlights on the object surface. In the proposed method, the highlight regions are detected using the analytic complex wavelet transform. Then based on an iterative regularization model, the missing fringe pattern can be recovered by filling the corresponding highlight area in the complex wavelet domain. Simulation and experiments were carried out using synthetic and real objects in different lighting conditions. Experimental results show that the proposed algorithm can accurately reconstruct the 3D model of the object with highlight regions.