Yapay Zeka Hızlı Bakış & Çıkarımlar
AI InsightThe performance of the method is evaluated with simulation results and real data, and visually compared to popular unwrapping methods.
📌 3 Maddede Bu Makale
Abstract
The method of reflecting sinusoidal phase-shifted patterns to the surface, based on the demodulation technique of phase information, has been a popular method to obtain 3-D surface depth using 2-D images. The phase information that is extracted with this method is wrapped; so it must be unwrapped. Even though the phase information is unwrapped, there will be some errors because of the possibility of non-sinusoidal characteristics of phase patterns, surface discontinuities, low sample rates, and technical handicaps (poor calibration and hardware malfunctions, and so on). To deal with these errors resulting from the phase unwrapping process, there are some computationally expensive and complex methods that have been presented. In this paper, a fast and low complex local adaptive phase correction algorithm based on the four-step phase shifting method is implemented. The method is firstly validated by using synthetic data. After the validation process, an optic test system is realized, and a few experiments are performed by using physical real data. For the optical system used to physically acquire the data, a lookup table-based calibration technique has also been developed to obtain accurate surface phase information. The performance of the method is evaluated with simulation results and real data, and visually compared to popular unwrapping methods.