Improvement of Algorithms in the Particle Tracking Velocimetry

Ph.D. Candidate: Mr. Purushottam Shrestha

Particle tracking velocimetry (PTV)/ Particle image velocimetry (PIV) is a multi-disciplinary research field. The spectrum of research includes bio-medical, electrical, electronics, flow visualization, heat flow visualization, and nuclear engineering. The PTV quantifies the displacement of individual dynamic particles between two or more image frames. Several experimental engineers involved in the PIV measurement of fluid flows are well aware of the importance of both extraction and tracking algorithms. The extraction algorithm is for identifying individual particles and detecting their respective centers from the images coded in gray scale levels.

The particle images generated in flow visualization experiments are not even in size, shape, and brightness. The extraction of individual particles is a problem for the accuracy of particle tacking. Apart from this, the particle images are not free from partial or entire overlap with their neighbors. Similarly there are added random noises superposed on the background image. There seem to be no effective algorithm applicable to any type of experimental particle images with successful results. The research will be mainly concentrated to develop improved algorithm for the extraction of the particle from image and to track it