Register of ankle angle kinematics data using Functional Data Analysis for parachute landing

The biomechanical data encountered a significant and common difficulty related to event-time alignment due to the fact that each motion capture (MoCap) trial exhibited varying durations to complete the tasks involved. To address this challenge, registration, which is essential for Functional Data Analysis (FDA), needed to be further examined in detail. The primary goal of this study was to evaluate the effectiveness of using both landmark and continuous registration methods on ankle angle data collected during the parachute landing fall activity. In total, twenty-one Malaysian forces participated in this experiment, which included three professional parachutists and eighteen amateurs. Their movements were recorded through two Go-Pro Hero 4 cameras, and the Quintic Biomechanics Software v26 was employed to extract the ankle angle data. The process of registration could be carried out in two distinct ways: by utilizing landmark points and through continuous registration. Landmark registration was performed initially, followed by continuous registration as an additional technique aimed at enhancing the alignment of the curves. Notably, the effectiveness of the registration process was assessed using the Root Mean Square Error (RMSE) test, which revealed that the ankle angle curves were aligned with considerable effectiveness when applying the landmark registration in conjunction with the continuous registration technique. The results yielded minimal RMSE values for both amateur and professional categories, recorded at 8.359 and 9.642, respectively. Moreover, the implementation of these registration techniques notably improved the accuracy of the average measurements for ankle angles.

Text 124142.pdf - Published Version Available under License Creative Commons Attribution. Download (1MB) Official URL or Download Paper: https://pphmjopenaccess.com/fejms/article/view/412...

Actions (login required)

View Item