Research on video target tracking technology based on improved SIFT algorithm

Research on video target tracking technology based on improved SIFT algorithm
复制标题

DOI：
10.1117/12.2265460
复制
发表时间：
2017-01
期刊：
--
影响因子：
--
通讯作者：
Zhemin Zhuang;Zhijie Guo;Ye Yuang
Zhemin Zhuang;Zhijie Guo;Ye Yuang
中科院分区：
其他
文献类型：
--
作者：
Zhemin Zhuang;Zhijie Guo;Ye Yuang

文献摘要

相似文献

提出了一种基于改进SIFT（ScaleInvariantFeatureTransform，尺度不变特征变换）算法的目标跟踪算法。为了提高实时性，对SIFT的处理邻域进行了改进，降低了计算复杂度，并将SIFT向量的维数从128降为40。仿真和实验表明，该算法计算复杂度低，跟踪精度高，鲁棒性好。
A novel target tracking algorithm based on improved SIFT (Scale Invariant Feature Transform，SIFT） algorithm is proposed in this paper. In order to improve real-time performance, the processing neighborhood of SIFT has been improved to decrease the complexity of calculation, and the dimension of the SIFT vector is set from 128 to 40. Simulations and experiments show this improved algorithm brings us low computation complexity and high tracking accuracy and robustness.