CAJ | 학술논문

传统的目标跟踪算法需要人为标定跟踪区域，且受到漂移问题的影响。为了解决这些困难，针对人体跟踪问题，提出了一种新的基于检测的跟踪算法。为了减少漏跟踪，使用了多个检测算子，用来定位多个身体部位，将其检测结果映射到一个相同的身体区域。为了适应快速运动的目标，使用KLT跟踪和凝聚聚类将检测窗口连接起来形成人体轨迹。实验结果表明：使用多个检测算子明显地提高了跟踪性能；KLT跟踪对于快速运动目标具有适应能力。该算法基本满足实时性。
전통적목표근종산법수요인위표정근종구역，차수도표이문제적영향。위료해결저사곤난，침대인체근종문제，제출료일충신적기우검측적근종산법。위료감소루근종，사용료다개검측산자，용래정위다개신체부위，장기검측결과영사도일개상동적신체구역。위료괄응쾌속운동적목표，사용KLT근종화응취취류장검측창구련접기래형성인체궤적。실험결과표명：사용다개검측산자명현지제고료근종성능；KLT근종대우쾌속운동목표구유괄응능력。해산법기본만족실시성。
The traditional object tracking algorithms require manually annotated tracking area,and suffer from the problem of drift.To address these difficulties,the problem of person tracking was focused on,and a new detection-based tracking algorithm was proposed.To reduce failure in tracking,multiple detectors to locate multiple body parts were employed,and then their detection results were mapped to a common body area.To adapt for the quickly moving objects,the KLT tracker and agglomerative clustering for linking the detection windows to form person body trajectories was employed.The experimental results reveal that using multiple detectors improves the tracking performance significantly,and the KLT tracker is adaptable for quickly moving objects.This algorithm is nearly real-time.