摘要
传统的模板匹配跟踪算法存在运算量大和实时性差的缺陷,限制了它的应用范围。针对这一问题,在已有的基于位置预测相关跟踪算法的基础上,提出了利用图像分离块操作对相关跟踪算法进行改进,进一步减少相关跟踪算法的计算量,以满足系统的实时性要求。对改进算法进行了模拟仿真实验验证。仿真结果表明:该改进算法不仅在很大程度上减少了计算量,提高了跟踪系统的实时性,而且能够有效地减少随机噪声的影响,使得跟踪更加快速准确。
The application scope of traditional template matching tracking algorithm is limited because of its large computation load and poor real-time performance. In order to overcome this problem, it shall be optimized to meet the real-time requirement. A modified real-time imaging tracking algorithm based on the available position prediction relevant tracking algorithm is presented by utilizing the image block processing to decrease the calculation toad of traditional correlation tracking algorithm and improve the real-time performance. The modified algorithm was simulated and verified in the experiment. The result of simulation indicates that the presented algorithm decreases the matching calculation load and improves real-time performance, reduces the effect of the stochastic noise and makes tracking faster and accurate.
出处
《应用光学》
CAS
CSCD
北大核心
2009年第2期195-198,共4页
Journal of Applied Optics
基金
国家自然科学基金资助项目(60572079)
关键词
模板匹配
分离块操作
实时跟踪
位置预测
template matching
separation block processing
real-time tracking
position prediction
