1 PIV error estimation by image matching Andrea Sciacchitano 1 , Bernhard Wieneke 2 and Fulvio Scarano 1 1 Aerospace Engineering Department, Delft University of Technology, Delft 2629 HS, The Netherlands 2 LaVision GmbH, Göttingen D-37081, Germany E-mail: a.sciacchitano@tudelft.nl Abstract A novel method is presented to estimate the error of PIV data. The approach is a-posteriori, i.e. the error of the measured velocity field is estimated using the velocity field itself as input along with the original images. The principle of the method relies on the concept of super-resolution: the image pair is matched according to the cross-correlation analysis and the residual distance between matched particle image pairs (particle disparity vector) due to incomplete match between the two exposures is measured. The ensemble of disparity vectors within the interrogation window is analyzed statistically. The dispersion of the disparity vector returns the estimate of the random error, whereas the mean value of the disparity indicates the occurrence of a systematic error. The validity of the working principle is first demonstrated via Monte Carlo simulations. Two different interrogation algorithms are considered, namely the cross-correlation with discrete window offset and the multi-pass with window deformation. In the simulated recordings, the effects of particle image displacement, its gradient, out-of-plane motion, seeding density and particle image diameter are considered. In all cases a good agreement is retrieved, indicating that the error estimator is able to follow the trend of the actual error with satisfactory precision. Experiments where time-resolved PIV data is available are used to prove the concept in realistic measurement conditions. In this case the “exact” velocity field is unknown, however a high-accuracy estimate is obtained with an advanced interrogation algorithm that exploits the redundant information of highly temporally oversampled data. The image-matching estimator returns the instantaneous distribution of the estimated velocity measurement error. The spatial distribution compares very well with that of the actual error with maxima in the highly sheared regions and in the 3D turbulent regions. The high level of correlation between the estimated error and the actual error indicates that this new approach can be utilized to directly infer the measurement uncertainty from PIV data. A procedure is shown where the results of the error estimation are employed to minimize the measurement error by selecting the optimal interrogation window size. Keywords PIV, error estimation, uncertainty, image matching, super-resolution, a-posteriori error analysis 1. Introduction Digital particle image velocimetry (PIV) is nowadays an established and reliable flow diagnostic tool capable of measuring velocity fields in two- and three-dimensional domains. Several works have been focused on PIV measurement errors. Fincham and Spedding (1997) distinguished two major forms of errors in digital PIV, namely the mean-bias error and the root-mean-square (RMS) error of the remaining fluctuations (or precision error). The bias error mostly occurs when the particle image size relative to that Confidential: not for distribution. Submitted to IOP Publishing for peer review 27 November 2012