Noise reduction of temporal phase unwrapping for 3D shape reconstruction using a quality map

Abstract

Nowadays, it is possible to find several works on surface reconstruction with optical techniques that employ quality maps, all of them using laboratory cameras with several functions such as manual gain control, low value of white noise, etc. In this work we analyze the possibility to obtain a good 3D shape reconstruction using a webcam, with the aim to develop a low cost portable system. In this work, quality maps are designed to be implemented in the temporal phase unwrapping algorithm with fringe projection for 3D shape reconstruction. The combined techniques allow the noise suppression and the study of illumination areas obtained through fringe projection. In particular, threshold value of quality maps allows to select areas of high visibility in which the temporal phase unwrapping algorithm can be successfully applied. The effect of the position of projector, camera, and sample into the optical setup are considered in order to obtain the optimum fringe density and to suppress Moiré or sampling effects in the area of interest. A low pass spatial filter, applied to the temporal phases, suppresses residual noise. The implemented camera calibration process demonstrates that image distortion at optimum spatial locations is negligible. The final 3D shape reconstructions show that the noise usually generated by shadows and dark areas of the specimen under study can be successfully eliminated.

Holographic and weak-phase projection system for 3D shape reconstruction using temporal phase unwrapping

“…4 shows the result obtained by projecting fringe patterns using kinoforms, in this figure a calibration process 17,18 was implemented to recover its original scale eliminating the spherical aberration. As we can observe the technique is able to recover the shape, but is not good enough to obtain resolutions less than a millimeter.…”