Method and system for automatically counting moving targets in fitness video

CN121053162APending Publication Date: 2025-12-02HARBIN FINANCE UNIV
0 Cites 2 Cited by

Patent Information

Application Number
CN202510933564.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-08
Publication Date
2025-12-02

Smart Images

  • Figure CN121053162A_ABST
    Figure CN121053162A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of target recognition, in particular to an automatic counting method and system for moving targets in a fitness video, and the method comprises the following steps: screening regions which accord with the relative position relation of hip joints, knee joints and ankle joints in continuous image frames of the fitness video, and extracting the coordinates of the center of mass; according to the method, the areas conforming to the relative position relation of the hip joint, the knee joint and the ankle joint are screened in the continuous image frames, the human body feature points are positioned by means of the centroid coordinate extraction mode, the recognition precision of the key skeleton structure in the target recognition stage can be improved, and interference caused by posture changes is reduced. On the basis, through sequential association of target joint centroid coordinates between continuous image frames, a motion path is constructed, a dynamic updating mechanism of a state vector is introduced, coordinate positions and speeds are used as state description quantities, and a smooth trajectory sequence with higher dynamic consistency is constructed in combination with residual feedback of a measured value and a predicted value of a current frame.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Cited By

  • Anti-interference industrial visual counting method and device based on spatio-temporal context perception

    CN122199550A

  • Anti-interference industrial vision counting method and device based on spatio-temporal context perception

    CN122199550B