基于Canny-Harris特征点的图像目标尺度方向跟踪算法

Image target scale direction tracking algorithm based on Canny-Harris feature points

传统图像目标尺度方向跟踪算法无法实时更新所跟踪目标外观模型,导致在光照等外界因素影响下跟踪误差较大。提出一种基于Canny-Harris特征点的图像目标尺度方向跟踪算法。计算各个图像块属于目标的实际概率,构建目标外观模型;建立当前帧中目标关于位置的对应置信度图,分析该置信度图获取当前帧中目标的最优状态;基于Canny-Harris特征点设计更新目标外观模型策略,在线实时更新目标外观。实验结果证明了该算法在图像噪声比为50dB~110dB的范围内跟踪误差较小,实现了算法性能的提升。

The traditional image target scale direction tracking algorithm cannot update the appearance model of the tracked target in real time,which leads to a large tracking error under the influence of external factors such as illumination.Therefore,an image target scale direction tracking algorithm based on Canny-Harris feature points is proposed.The actual probability of each image block belonging to the target is calculated,the target appearance model is constructed,and the corresponding confidence graph of the position of the target in the current frame is established.The confidence graph is analyzed to obtain the optimal state of the target in the current frame.The strategy of updating target appearance model is designed based on Canny-Harris feature points,and the target appearance is updated online in real time.Experiment results show that the tracking error of the algorithm is small in the range of 50dB~110dB image noise ratio,and the algorithm performance is improved.