| 摘要 |
The present invention relates to a fringe image phase distribution analysis technique that performs one-dimensional discrete Fourier transform using temporal intensity information or spatial intensity information to calculate the phase distribution of the fringe image. The contrast (signal-to-noise ratio) of the acquired fringe image is reduced due to extreme low or high reflectance of the object to be measured and a large error occurs in the analysis result of a phase distribution, or a large measurement error occurs when an error is included in the amount of phase-shift due to environmental vibration during measurement or depending on the performance of the phase-shift device. Therefore, it is necessary to improve the analysis accuracy of the phase distribution. A plurality of phase-shifted moire fringe images is generated from high-dimensional intensity data by a thinning-out (down-sampling) process and an image interpolation process, and the phase distribution of the moire fringe is calculated by two-dimensional or three-dimensional discrete Fourier transform. In addition, the phase distribution of thinned-out is added to calculate the phase distribution of an original fringe image. Since high-dimensional intensity information which is present in both spatio-domain and temporal-domain is used, phase distribution analysis is less likely to be affected by random noise or vibration. In addition, even when measurement conditions are poor, it is possible to perform phase distribution analysis with high accuracy. |
