Browse/search for people

Publication - Mr Alessandro Masullo

    Improvement of PIV dynamic range in the presence of velocity gradients using multiple correlation peak analysis and self-adaptive windows

    Citation

    Masullo, A & Theunissen, R, 2016, ‘Improvement of PIV dynamic range in the presence of velocity gradients using multiple correlation peak analysis and self-adaptive windows’. in: The International Symposia on Applications of Laser Techniques to Fluid Mechanics.

    Abstract

    A novel algorithm to analyse PIV images in case of strong gradients is proposed. The proposed methodology allows the detection of multiple peaks in a correlation map and automatically collocates vectors according to the match of each peak with the particles that constitute it. The algorithm is constituted by two main parts: an automatic peak detection system based on the histogram of the cross-correlation map and the signal to noise ratio, and a peak matching system that identifies which particles participate to which peak, by use of a sub-correlation window and a defined matching function. An adaptive windows size based on the flow predictor and the quarter rule is also implemented to help the resolution of gradients. Synthetic and experimental results are proposed and show a strongly reduced error in the measurement and a velocity dynamic range improved up to a factor of 4.

    Full details in the University publications repository