Post-processing of particle tracking data for phenomenological depiction of weak bed-load sediment transport

Image processing offers unique capabilities for phenomenological analysis of bed-load sediment transport processes. However, a number of issues emerge when post-processing the raw data obtained from images. The present study aims at proposing strategies to handle such issues, towards appropriate measurement and statistical analysis of particle kinematics under weak bed-load transport. Two major issues are addressed here: (i) definition of motion and stillness states for tracked particles and (ii) possibility/need to introduce a minimum threshold for hop length (i.e., for motion events to be considered ‘significant’). Analyses are supported by tracking data from a recent experiment where a CCD camera placed over a pressurized channel recorded the motion of white particles released over a fixed, rough black bed. Particles were tracked from image to image and some different post-processing algorithms were compared in terms of resulting statistical properties of particle kinematic quantities. A general recommendation is strongly proposed, according to which clear description of used post-processing methods is needed to ensure possibility of comparison among data sets from different scholars.