Improving the Delaunay tessellation particle tracking algorithm in the three-dimensional field

•PTV algorithm is preferable in the reconstruction of 3D particle flow.•We improve the Delaunay tessellation PTV and name it 3DT-PTV.•3DT-PTV is structurally compact and highly free of algorithmic assumptions.•A testing process is realized by introducing synthetic flows with perturbations.•The superiority of 3DT-PTV over the traditional version is confirmed.

Particle tracking velocimetry (PTV) algorithms match particles across two consecutive frames corresponding to a certain flow pattern. PTV based on Delaunay tessellation (DT-PTV) has a succinct structure and depends minimally on algorithmic assumptions. This study proposes several methods for improving DT-PTV performance in the three-dimensional field. The improved version, called 3DT-PTV, is tested using synthetic flows with various parameters and under difficult circumstances. Results show that 3DT-PTV performs better than the classical version in addressing flows with a noticeable ratio of particles without match and those with an excessive ratio of the inter-frame particle displacement to the inter-particle distance.