Aiming at the problems of high efficient feature extracting and dealing with the model drift,a kernel correlation filter tracking algorithm based on saliency detection is proposed. The color feature and the histogram of oriented gradient are weightedly merged,and features weights can be adjusted adaptively. For dealing with the mod-el drift,and inspired by biological vision mechanism, target salient region is obtained by sampling in the region through the visual saliency algorithm. In that way, the sampling in the area is carried out to complete the global scope search which avoids local maximum. In addition, a model based on the key points is introduced to dealing with the problem that the target scale is changed. In order to verify the proposed algorithm effectiveness,it is com-pared with 5 kinds of excellent algorithms in recent years on 50 video sequences. Experimental results show that, the proposed algorithm can obtain better results in the success rate and the center position error,and what is more, the algorithm can effectively alleviate target model drift.%针对在线视觉跟踪中的高效特征提取以及模型漂移的问题,提出了一种基于显著性检测的核相关滤波器(KCF)跟踪算法.将颜色特征(CN)和方向梯度直方图(HOG)进行加权融合;并自适应地调节每种特征的权重.对于模型漂移问题,受生物视觉机制的启发,通过视觉显著性算法获得目标的显著区域;并在该区域内进行采样,实现了全局范围搜索,避免陷入局部极大值.此外,引入了一种基于关键点的模型来解决目标尺度固定的问题.为验证提出算法的有效性,在50个视频序列上与近年来的5种优秀算法进行了对比.实验结果表明,与以往算法相比,该算法在成功率和中心位置误差上都取得较好的效果;而且能有效地缓解目标模型漂移问题.
展开▼
