Affiliations: [a] School of Physical Education, Yanching Institute of Technology, Langfang, Hebei, China | [b] Physical Education Department, Shangqiu Institute of Technology, Shangqiu, Henan, China
Correspondence:
[*]
Corresponding author: Yushi Qie, School of Physical Education, Yanching Institute of Technology, Langfang, Hebei 065201, China. E-mail: [email protected].
Abstract: In order to solve the problems of low recognition efficiency, low recognition rate and large recognition error of traditional methods, an application method of artificial intelligence technology in athletes’ running foul recognition was proposed. Build the image acquisition model of sports athletes’ running foul, divide each frame of the image samples into static area and motion area, and get the motion direction estimation results; K-means in the field of artificial intelligence is used to cluster the characteristics of sports athletes’ rush foul action, and LLE algorithm is used to reduce the dimension of features; The background subtraction method is used to detect the foul target of rush, and the Bayesian algorithm is used to construct the recognition model of sports athletes’ foul of rush, which is used to identify the foul target. The experimental results show that the recognition rate of this method has reached more than 72%, and continues to increase, and the recognition error is only 2%, which effectively improves the recognition rate and reduces the recognition error, which is feasible and effective.
Keywords: Artificial intelligence, foul identification, K-means clustering, feature dimensionality reduction, LLE algorithm, background subtraction, static area, dynamic area
