本文已被：浏览 1476次 下载 618次
投稿时间：2011-07-06
投稿时间：2011-07-06
中文摘要: 探讨了稀疏投影表示法在目标检测跟踪系统中的应用.通过一范数(L1)正则化最小二乘算法实现目标特征在模板矩阵的稀疏投影,并根据投影差值最小的目标特征找到最优跟踪状态估计,最后更新模板矩阵以适应目标变化.该方法将基于贝叶斯框架的状态预测和视频采集模板相结合得到最优跟踪轨迹.实验结果表明,该方法能够达到很好的跟踪效果.
Abstract:The application of sparse representation in target tracking system is discussed． Thesparse representation of target feature is achieved by solving an L1 regularized least squares problem，then the candidate with the smallest target template projection error is chosen as the tracking result．The templates are updated to adapt to the change of the object of tracking． Optimal tracking resultsare obtained through combining the Bayes framework based hypothesis with the templates of image sequences． Experiments show that the method is effective．
文章编号：20120414 中图分类号： 文献标志码：
基金项目:
| 作者 | 单位 | |
| 邵洁 | 上海电力学院计算机与信息工程学院, 上海200090 | shaoj．shiep@ gmail． com． |
| Author Name | Affiliation | |
| SHAO Jie | School of Computer and Information Engineering,Shanghai University of Electric Power,Shanghai 200090,China | shaoj．shiep@ gmail． com． |
引用文本：
邵洁.稀疏投影在目标跟踪中的应用[J].上海电力大学学报,2012,28(4):357-360.
SHAO Jie.Target Tracking With Sparse Representation[J].Journal of Shanghai University of Electric Power,2012,28(4):357-360.
邵洁.稀疏投影在目标跟踪中的应用[J].上海电力大学学报,2012,28(4):357-360.
SHAO Jie.Target Tracking With Sparse Representation[J].Journal of Shanghai University of Electric Power,2012,28(4):357-360.