Qian Kemao

Windowed Fourier transform for fringe pattern analysis

Fringe patterns in optical metrology systems need to be demodulated to yield the desired parameters. Time-frequency analysis is a useful concept for fringe demodulation, and a windowed Fourier transform is chosen for the determination of phase and phase derivative. Two approaches are developed: the first is based on the concept of filtering the fringe patterns, and the second is based on the best match between the fringe pattern and computer-generated windowed exponential elements. I focus on the extraction of phase and phase derivatives from either phase-shifted fringe patterns or a single carrier fringe pattern. Principles as well as examples are given to show the effectiveness of the proposed methods.

Phase error analysis and compensation for nonsinusoidal waveforms in phase-shifting digital fringe projection profilometry

The nonlinear intensity response of a digital fringe projection profilometry (FPP) system causes the captured fringe patterns to be nonsinusoidal waveforms and leads to an additional phase measurement error for commonly used three- and four-step phase-shifting algorithms. We perform theoretical analysis of the phase error owing to the nonsinusoidal waveforms. Based on the derived theoretical model, a novel and simple iterative phase compensation algorithm is proposed to compensate the nonsinusoidal phase error. Experiments show that the proposed algorithm can be used for effective phase error compensation in practical phase-shifting FPP.

Windowed Fourier transform for fringe pattern analysis: theoretical analyses.

A windowed Fourier ridges (WFR) algorithm and a windowed Fourier filtering (WFF) algorithm have been proposed for fringe pattern analysis and have been demonstrated to be versatile and effective. Theoretical analyses of their performances are of interest. Local frequency and phase extraction errors by the WFR and WFF algorithms are analyzed in this paper. Effectiveness of the WFR and WFF algorithms will thus be theoretically proven. Consider four phase-shifted fringe patterns with local quadric phase [c(20)=c(02)=0.005 rad/(pixel)(2)], and assume that the noise in these fringe patterns have mean values of zero and standard deviations the same as the fringe amplitude. If the phase is directly obtained using the four-step phase-shifting algorithm, the phase error has a mean of zero and a standard deviation of 0.7 rad. However, when using the WFR algorithm with a window size of sigma(x)=sigma(y)=10 pixels, the local frequency extraction error has a mean of zero and a standard deviation of less than 0.01 rad/pixel and the phase extraction error in the WFR algorithm has a mean of zero and a standard deviation of about 0.02 rad. When using the WFF algorithm with the same window size, the phase extraction error has a mean of zero and a standard deviation of less than 0.04 rad and the local frequency extraction error also has a mean of zero and a standard deviation of less than 0.01 rad/pixel. Thus, an unbiased estimation with very low standard deviation is achievable for local frequencies and phase distributions through windowed Fourier transforms. Algorithms applied to different fringe patterns, different noise models, and different dimensions are discussed. The theoretical analyses are verified by numerical simulations.

Quality-guided phase unwrapping technique: comparison of quality maps and guiding strategies

Quality-guided phase unwrapping is a widely used technique with different quality definitions and guiding strategies reported. It is thus necessary to do a detailed comparison of these approaches to choose the optimal quality map and guiding strategy. For quality maps, in the presence of noise, transform-based methods are found to be the best choice. However in the presence of discontinuities, phase unwrapping is itself unresolved and hence quality-guided phase unwrapping is not sufficient. For guiding strategies, classical, two-section, and stack-chain guiding strategies are chosen for comparison. If accuracy is the foremost criterion then the classical guiding strategy with a data structure of indexed interwoven linked list is best. If speed is of essence then the stack-chain guiding strategy is the one to use.

Windowed Fourier-filtered and quality-guided phase-unwrapping algorithm.

We propose a windowed Fourier-filtered and quality-guided phase-unwrapping algorithm that is an extension and improvement of our previous phase-unwrapping algorithm based on windowed Fourier transform [Opt. Laser Technol.37, 458 (2005), Key Eng. Mater.326-328, 67 (2006). First, the filtered amplitude is used as a real-valued quality map, rather than a binary mask, which makes the phase-unwrapping algorithm more tolerant to low-quality regions in a wrapped-phase map, and the process is more automatic. Second, the window size selection is considered, which enables the algorithm to be adapted to tackle different phase-unwrapping problems. A large window size is useful for removing noise, building long barriers along phase discontinuities, and identifying invalid regions, while a small window size is useful for preserving local features, such as small regions and high-quality narrow channels. Eight typical examples in Ghiglia and Pritts excellent book Two-Dimensional Phase Unwrapping: Theory, Algorithm and Software (Wiley, 1998) are used to evaluate the proposed algorithm. The proposed algorithm is able to unwrap all these examples successfully. The windowed Fourier ridges algorithm, another algorithm based on windowed Fourier transform, is also tested and found to be useful in building barriers along phase discontinuities.

Fringe pattern denoising using coherence-enhancing diffusion

Electronic speckle pattern interferometry is one of the methods measuring the displacement on object surfaces in which fringe patterns need to be evaluated. Noise is one of the key problems affecting further processing and reducing measurement quality. We propose an application of coherence-enhancing diffusion to fringe-pattern denoising. It smoothes a fringe pattern along directions both parallel and perpendicular to fringe orientation with suitable diffusion speeds to more effectively reduce noise and improve fringe-pattern quality. It is a generalized work of Tangs et al.s [Opt. Lett.33, 2179 (2008)] model that only smoothes a fringe pattern along fringe orientation. Since our model diffuses a fringe pattern with an additional direction, it is able to denoise low-density fringes as well as improve denoising effectiveness for high-density fringes. Theoretical analysis as well as simulation and experimental verifications are addressed.

Frequency guided methods for demodulation of a single fringe pattern

Phase demodulation from a single fringe pattern is a challenging task but of interest. A frequency-guided regularized phase tracker and a frequency-guided sequential demodulation method with Levenberg-Marquardt optimization are proposed to demodulate a single fringe pattern. Demodulation path guided by the local frequency from the highest to the lowest is applied in both methods. Since critical points have low local frequency values, they are processed last so that the spurious sign problem caused by these points is avoided. These two methods can be considered as alternatives to the effective fringe follower regularized phase tracker. Demodulation results from one computer-simulated and two experimental fringe patterns using the proposed methods will be demonstrated.

Sequential demodulation of a single fringe pattern guided by local frequencies

A simple but effective approach for the demodulation of a single fringe pattern is proposed. The phase with an undetermined sign is directly obtained by taking the arccosine value of a preprocessed fringe pattern. The local frequencies, also with an undetermined sign, are then estimated by local matching. The sign ambiguity is then removed simply by forcing the continuity of the local frequencies. The priority of sign determination is guided by the value of total local frequency (fringe density) so that the critical points are processed last. The proposed approach is verified by successful demodulation of a simulated fringe pattern and two experimental fringe patterns.

Comparative analysis on some filters for wrapped phase maps.

Some effective filtering methods for wrapped phase maps, a regularized phase-tracking method (RPT) without the regularization term, a multiple-parameter least-square method (MPLS), a windowed Fourier ridges method (WFR), an autocorrelation function method (ACF), and a sine/cosine average filter (SCAF), are analyzed in order to establish their transversal relationship. The analysis shows that principles of the RPT, MPLS, WFR, and ACF are equivalent and the SCAF also leads to the WFR by some extension, which elegantly unifies all these methods for filtering unwrapped phase maps.

Real-time 2D parallel windowed Fourier transform for fringe pattern analysis using Graphics Processing Unit

In optical interferometers, fringe projection systems, and synthetic aperture radars, fringe patterns are common outcomes and usually degraded by unavoidable noises. The presence of noises makes the phase extraction and phase unwrapping challenging. Windowed Fourier transform (WFT) based algorithms have been proven to be effective for fringe pattern analysis to various applications. However, the WFT-based algorithms are computationally expensive, prohibiting them from real-time applications. In this paper, we propose a fast parallel WFT-based library using graphics processing units and computer unified device architecture. Real-time WFT-based algorithms are achieved with 4 frames per second in processing 256x256 fringe patterns. Up to 132x speedup is obtained for WFT-based algorithms using NVIDIA GTX295 graphics card than sequential C in quad-core 2.5GHz Intel(R)Xeon(R) CPU E5420.