Skill and tactic diagnosis for table tennis matches based on artificial neural network and genetic algorithm.
ABSTRACT Due to the complexity, multiplicity and randomness of table tennis matches, the paper presents skill and tactic diagnostic model for table-tennis matches of elite athletes with artificial neural network and genetic algorithm. A back propagation network is used to build basic structure of the model and genetic algorithm is established to optimize the connection weights and threshold values of the neural network to improve the prediction precision and congestion performance. The application results show that it is an effective tool to provide decision support for table-tennis players.
[Show abstract] [Hide abstract]
ABSTRACT: Due to complexity, multiplicity and randomness of table tennis matches, it is necessary to develop some skill and tactic diagnostic models especially an efficient scoring model to get some skill and tactics data for table tennis matches. A scoring model is developed for table tennis matches based on video image processing technology. An input match video is used to analysis the skill and tactic of the players. Once the player scored more than some criteria points constantly, it records this particular event either in the notebook or in database to arouse his attention. By playing the fragment of the record video either slow or fast as he likes, the player can know not only the reason why he lose the points, but also know the skill and tactic of the opponent. Besides it gives a dynamic grow of histogram for every score. The proposal improves the efficiency and quality of the matches. Moreover it provides an efficient guidance in the future's plan making, practice or competition for coachers and athletes.