Based on the continuous development of computer vision and graphics technology, pattern recognition and image processing technology, intelligent video analysis technology combines these related technologies to establish a mapping relationship between image description and surveillance image. Intelligent video analysis technology has a wide range of applications, such as in military and economic fields. In the process of badminton training, when the trainer explains the action essentials for the athletes, the video replay and analysis management methods are less difficult, lack of science and intuition and it is difficult to meet the trainer's demand for sports performance evaluation. The analysis and recognition process of badminton video are studied. Based on particle filter prediction, adaptive threshold algorithm is used to achieve adaptive separation of moving targets. The particle filter technology is used to track the trajectory of the athletes and the motion model is constructed. Based on this, the position of the important joint points in the subsequent motion video frames is predicted, so as to track the subsequent motion video frames. When the motion recognition in the badminton training video is used, the conditional random field method is mainly used, which makes the motion recognition rate higher and the false separation rate lower.
© Copyright 2019 IOP Conf. Series: Journal of Physics: Conf. Series 1213. Alle Rechte vorbehalten. / Übersetzt mit https://www.DeepL.com/Translator