Abstract.
This paper proposes an artificial neural network (ANN) method to effectively detect spurious velocity vectors in a velocity field measured by particle image velocimetry (PIV). The neural network is a recurrent network referred to as a cellular neural network (CNN). The method is compared with the local-median method to remove measurement outliers. Both artificially generated velocity fields containing known errors and actual experimental data were used to study the performance of these methods. The influences of the velocity gradient and the error percentage are discussed. The CNN model was shown to be more efficient for removal of erroneous vectors.
Similar content being viewed by others
Author information
Authors and Affiliations
Additional information
Electronic Publication
Rights and permissions
About this article
Cite this article
Liang, .D., Jiang, .C. & Li, .Y. Cellular neural network to detect spurious vectors in PIV data. Exp Fluids 34, 52–62 (2003). https://doi.org/10.1007/s00348-002-0530-8
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/s00348-002-0530-8