大连民族学院学报
大連民族學院學報
대련민족학원학보
JOURNAL OF DALIAN UNIVERSITY FOR NATIONAL MINORITIES
2015年
3期
259-264
,共6页
张丹%陈兴文%赵姝颖%程立英
張丹%陳興文%趙姝穎%程立英
장단%진흥문%조주영%정립영
TLD算法%随机森林%目标跟踪%LK光流法
TLD算法%隨機森林%目標跟蹤%LK光流法
TLD산법%수궤삼림%목표근종%LK광류법
TLD algorithm%random forest%target tracking%LK optical flow method
结合正负样本相互作用思想和随机森林算法构建检测器,融合基于LK光流法的跟踪器,提出一种基于TLD( Tracking Learning Detecting)的随机森林长期目标跟踪方法。将该方法与Mean-Shift算法、TLD算法进行对比,结果表明该算法能很好应对目标丢失、遮挡情况,准确率在93%以上。在多种情况下对该方法进行实验验证,可实现刚性物体和非刚性物体在复杂背景下的长时间精确跟踪。
結閤正負樣本相互作用思想和隨機森林算法構建檢測器,融閤基于LK光流法的跟蹤器,提齣一種基于TLD( Tracking Learning Detecting)的隨機森林長期目標跟蹤方法。將該方法與Mean-Shift算法、TLD算法進行對比,結果錶明該算法能很好應對目標丟失、遮擋情況,準確率在93%以上。在多種情況下對該方法進行實驗驗證,可實現剛性物體和非剛性物體在複雜揹景下的長時間精確跟蹤。
결합정부양본상호작용사상화수궤삼림산법구건검측기,융합기우LK광류법적근종기,제출일충기우TLD( Tracking Learning Detecting)적수궤삼림장기목표근종방법。장해방법여Mean-Shift산법、TLD산법진행대비,결과표명해산법능흔호응대목표주실、차당정황,준학솔재93%이상。재다충정황하대해방법진행실험험증,가실현강성물체화비강성물체재복잡배경하적장시간정학근종。
In this paper, we propose a target tracking method based on Tracking Learning Detec-ting ( TLD) random fores by using the detector constructed by the ideas of the interaction be-tween positive and negative samples and random forest algorithm, and tracker based on LK opti-cal flow method. This method is performed the comparison with the Mean Shift algorithm and TLD method. The results show that the algorithm can have the strong robustness to target lost, target occlusion, and the accuracy rate is more than 93%. The experiment results in many cases verify that this method can achieve a long time accurate tracking for rigid and non-rigid object in complex background.