An EMG Knowledge-Based System for Leg Strength Classification and Vertical Jump Height Estimation of Female Netball Players

Umar Yahya, S. M. N. Arosha Senanayake, Daphne Lai

Abstract: This study proposes a framework for the design and implementation of a leg strength Knowledge-Based Classification System based on pattern sets obtained from the statistical analysis of Electromyography (EMG) features (Peak, IEMG, Peak time) extracted from eight different lower limb muscles during a single leg vertical jump test of female netball players. The system employs a novel classification algorithm that takes into account the test subject’s anthropometry profile (height, body mass) and age, with respect to pattern sets in the knowledge-base. The algorithm is also able to estimate the vertical jump height with which the test subject attained the corresponding EMG data set. Results indicate that the system is not only able to accurately classify a subject’s overall leg strength category (strength category weight ≥ 66.67 %), but also is able to reveal the individual muscle strength classification (strength category weight ≥ 62.5 %). The results offer conditioning coaches and trainers deeper insights into muscle-specific strength variations of players.