Comparison of Kinematic Data Obtained by Inertial Sensors and Video Analysis

Abstract

The human body is an articulated system that, among other movements, presents a set of essential information in gait on the first assessment of body condition. Due to its complexity, an in-depth analysis of this movement requires interpretations that involve all biomechanics. In this context, this work's general objective was to analyze and interpret kinematic data, responsible for describing the movement quantitatively by values of displacement, velocity, and acceleration of the body. For this, two types of data collection were performed, one through the filming of the individual in their gait movement, in which strategic posterior markings were made on the foot and knee in each frame of the video, using the MATLAB software, and the other through the MPU-92.65 accelerometer sensor associated with an Arduino microcontroller board and manipulated by MATLAB software. The objective of carrying out these two acquisition methods is to compare these analyzes and verify the results obtained. This study revealed that the two methods showed similar results when analyzing the knee speed curve. However, due to some movement artifacts, the same result cannot be obtained for the foot speed curve.