@article{4384,
  author = {Cheng Zhou},
  title = {Moving Object Detection and Tracking Technology Based on Hybrid Algorithm},
  journal = {Journal of Multimedia Processing and Technologies},
  year = {2025},
  volume = {16},
  number = {1},
  doi = {https://doi.org/10.6025/jmpt/2025/16/1/20-27},
  url = {https://www.dline.info/jmpt/fulltext/v16n1/jmptv16n1_3.pdf},
  abstract = {Object tracking is a hot topic in visual technology and is widely applied in scenarios such as intelligent
monitoring, autonomous driving, and robot visual perception. In recent years, with the sports industry’s rapid
development, tracking targets (balls and players) in complex sports scenes represented by basketball and
football has attracted increasing attention. This paper focuses on tracking targets (balls as single targets and
players as multiple targets) in competitive sports scenes like basketball and football. A small target detection
network based on multi-scale features and a triangulation algorithm is employed to fuse the two-dimensional
coordinates of the ball into three-dimensional coordinates. Additionally, a simplified motion model is proposed
for the non-linear motion of the ball, and a Kalman filter is used to obtain accurate and smooth threedimensional
tracking trajectories. The proposed method achieves 2D and 3D tracking accuracies of 0.81 and
0.92 on the basketball public dataset, respectively.},
}