International Journal of Applied Information Systems |
Foundation of Computer Science (FCS), NY, USA |
Volume 12 - Number 29 |
Year of Publication: 2020 |
Authors: Mohamed EL Aallaoui |
10.5120/ijais2020451856 |
Mohamed EL Aallaoui . Multiselectivity Spatio-Temporal Analysis for Motion and Velocity Capture. International Journal of Applied Information Systems. 12, 29 ( May 2020), 1-10. DOI=10.5120/ijais2020451856
This paper focuses on the development of a new 2D time wavelets family for image sequence that can capture different motion in image sequence, ranging from highly directional ones to fully isotropic ones. We propose a multiselectivity spatio-temporal analysis, defined by isotropic and multidirectional decomposition with different angular selectivity. The result is a dictionary of wavelet transform with different selectivity level, which provides a theoretical tool to select a best representation for motion and velocity. This representation was used to developer a velocity capture algorithm in image sequence. The Experimental results demonstrate the effectiveness of the proposed approach.