Multi-Cross-Correlation Method in Particle Image Velocimetry

Abstract

ABSTRACTA multi-cross-correlation method (MCCM) was developed in a particle image velocimetry (PIV) auto-processing system to reduce spurious vectors and improve accuracy of measurements. This technique is an improvement based on conventional cross-correlation method (CCM). Four typical neighboring interrogation windows were specified to be overlapped and calculated by MCCM. A high cross-correlation value is obtained in which many particle images match up with their corresponding spatially shifted partners, and small cross-correlation peaks due to interference of noises during experiments are reduced. Several parameters such as out-of-plane motions, particle size, and seeding density are considered for checking both MCCM and conventional PIV algorithms. The examination gives authenticity to the merits of MCCM for avoiding particles loss or mistaken velocity vectors.