• 제목/요약/키워드: LM-IHPDA (Linear-Multi-Target IHPDA)

검색결과 1건 처리시간 0.014초

적외선 영상에서 다수표적추적을 위한 LM-IHPDA 알고리듬 연구 (A Study of LM-IHPDA Algorithm for Multi-Target Tracking in Infrared Image Sequences)

  • 김태한;최병인;김지은;양유경;송택렬
    • 제어로봇시스템학회논문지
    • /
    • 제19권3호
    • /
    • pp.209-218
    • /
    • 2013
  • Military surveillance systems with electro-optical sensors can be used to track a number of targets efficiently and reliably. In MTT (Multi-Target Tracking), joint events in which different tracks share the same measurements may occur. Measurement-to-track assignment are computationally challenging because of the number of operations increases exponentially with number of tracks and number of measurements. IHPDA (Integrated Highest Probability Data Association) based on a 2D-Assignment technique can find an optimal solution for measurement to track one-to-one assignments for complex environments. In this paper, LM-IHPDA (Linear Multi-Target IHPDA) which does not need to form all feasible joint events of association and thus the computational load is linear in the number of tracks and the number of measurements. Simulation studies illustrate the effectiveness of this approach in an infrared image environment.