Optimal Decoding of Stripe Patterns with Window Uniqueness Constraint

摘要

We propose the optimal algorithm of decoding color stripe patterns generated from a pseudo-random sequence (PRS). One-dimensional correspondence is solved globally by a variant of dynamic programming matching (DPM) that imposes the window uniqueness of the PRS as a hard constraint. Our algorithm runs in linear time complexity with respect to the size of projected pattern and acquired image, which is as efficient as the conventional DPM. The performance of our method is demonstrated qualitatively and quantitatively using simulation data with known ground truth and real data.