Framework of knowledge-based system for practising long jumpers using movement recognition

Abstract

The long jump is one of the standard events in modern Olympic Games. It is a part of track and field. The long jump comprises of four phases: Approach run phase, Take-off phase, Flight phase and Landing phase. Each phase effects to construct the flight distance. If athletes execute right actions in each phase, it will increase their performance. Athletes need some coaches or experts to provide them suggestions. Nonetheless, there is a lack of experts in this field. In this paper, we demonstrate a new framework of a knowledge-based system for training long jumpers in order to support trainers or coaches in practising and monitoring the long jumping movement. The idea is to combine the knowledge engineering methods with computer vision techniques for constructing the expert system. The system will be able to capture movements of the long jump athletes in each phase, analyse and give the recommendation based on knowledge captured from experts.