基于级联特征的随机森林运动目标跟踪算法

A Moving Object Tracking Algorithm of Random Forest Based on Features Cascade

在运动目标检测与跟踪的过程中,实际环境下的目标旋转、目标遮挡以及光照变化等因素时常出现,而目标检测与跟踪的性能对这些复杂环境因素极为敏感,甚至易导致目标跟踪丢失。为了提高复杂环境下运动目标跟踪的鲁棒性和稳定性,提出一种基于级联特征的随机森林运动目标跟踪算法。该算法首先在保留目标关键信息的ASIFT特征中级联目标轮廓信息作为正样本,训练正样本生成随机森林分类后续序列图像特征;在此基础上将CamShift算法确定的目标搜索窗口中的非目标特征作为负样本,训练负样本并更新随机森林以改善特征分类性能;最后通过对正负样本特征加权计算目标搜索窗口质心以改善跟踪性能。实验结果表明,该算法能够在光照突变、遮挡以及目标旋转等复杂环境下有效地实现运动目标跟踪。

In the process of moving object detection and tracking,in actual scenarios where there usually exists complex environmental factors including object rotation,occlusion and illumination and so on,the performance of the object detection and tracking is easily affected by these complicated environmental factors,even they lead to the occurrence of the losses of object tracking.In order to improve the robustness and stability of moving object tracking under complex environment,we propose a random forest for moving object tracking algorithm based on features cascade.The ASIFT features of the moving object with retaining key information are cascaded with object contour information as positive sample set.Random forest which can be used to classify the features of the subsequent sequence images is realized through training the positive sample set.On the basis of cascading features,the features of non-object in the object search window determined by CamShift algorithm are taken as negative sample set,and the performance of feature classification is improved by training negative sample set which is used to update random forest.The centroid of the object search window is calculated by weight-based positive and negative sample sets to improve the tracking performance.The experiment indicates that the algorithm can effectively realize moving object tracking under complex environment such as illumination fluctuation,occlusion and object rotation.