Surface reconstruction method by photometric stereovision based on Oren-Nayar's model

摘要

This article proposes a surface reconstruction method by photometric stereovision based on Oren-Nayar's model. The classical photometric stereovision reconstructs the surface by the exploitation of Lambert's model. However Lambert's model does not take into account the complex geometric and radiometric phenomena such as masking, shadowing and inter-reflections between micro-facets on the surface. Thus this article will improve the photometric stereovision by using Oren-Nayar's model instead of Lambert's model. The photometric stereovision based on Oren-Nayar's model will calculate respectively the energies attenuated by the phenomena of masking, shadowing and the energies produced by the phenomenon of inter-reflections. At last, the surface reconstruction experiments show that the result obtained by this method is more precise than the one obtained by classical photometric stereovision.