A rapid non-iterative proper orthogonal decomposition based outlier detection and correction for PIV data

The present work proposes a novel method of detection and estimation of outliers in particle image velocimetry measurements by the modification of the temporal coefficients associated with a proper orthogonal decomposition of an experimental time series. Using synthetic outliers applied to two seque...

Full description

Saved in:
Bibliographic Details
Published in:Measurement science & technology 2016-12, Vol.27 (12), p.125303
Main Authors: Higham, J E, Brevis, W, Keylock, C J
Format: Article
Language:English
Subjects:
experimental fluid mechanics
image processing
outlier detection
particle image velocimetry
proper orthogonal decomposition
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The present work proposes a novel method of detection and estimation of outliers in particle image velocimetry measurements by the modification of the temporal coefficients associated with a proper orthogonal decomposition of an experimental time series. Using synthetic outliers applied to two sequences of vector fields, the method is benchmarked against state-of-the-art approaches recently proposed to remove the influence of outliers. Compared with these methods, the proposed approach offers an increase in accuracy and robustness for the detection of outliers and comparable accuracy for their estimation.
ISSN:0957-0233
1361-6501
DOI:10.1088/0957-0233/27/12/125303