期刊文献+

一种非刚性运动目标图像跟踪的改进算法 被引量:2

Improved algorithm of non-rigid object tracking
下载PDF
导出
摘要 针对玉米秸秆粉碎后抛撒运动的特点,提出一种改进的变形模板跟踪算法实现非刚性目标的运动跟踪,在两个方面对原有的算法进行了改进:1)选择适当的预测方法,减少遮挡物体对研究目标的影响。2)通过选择适当的特征区域把运动目标限制在局部区域,而不是对整帧图像进行处理。试验证明了该算法在目标跟踪的实现效率、精确性方面有一定的优势。 A tracking non-rigid objects algorithm based on deformable template is presented in this paper. Two impovements of the original algorithm are proposed: 1) Decreasing influence of occlusion objects on motion objects via suitable prediction methods; 2) Tracking motion objects are restricted in local regions rather than in whole image through selecting suitable characteristic regions. The experiments show that the algorithm has advantages in precision and efficiency in motion tracking.
出处 《农业工程学报》 EI CAS CSCD 北大核心 2004年第4期17-20,共4页 Transactions of the Chinese Society of Agricultural Engineering
基金 国家十五科技攻关课题(2001BA504B03)
关键词 变形模板 遮挡 非刚性目标跟踪 序列图像 deformable template occlusion non-rigid objects tracking image sequences
  • 相关文献

参考文献6

  • 1Zaki M. EMBOT: an enhanced motion-based object tracker[J]. Systems and Software,2004,69:149-158.
  • 2Horn B, Schunch B. Determining optical flow[J]. Artificial Intelligence, 1981,17:185-203.
  • 3Kass M, Witkin A P, Terzopoulos D. Snakes: active contour models[J]. Int J Computer Vis,1988,1:321-331.
  • 4Xu Chenyang, Prince J L. Snakes, shapes, and gradient vector flow[J]. IEEE Trans Image Proc,1998,7(3):359-369.
  • 5Won Kim, Choon Young Lee, Ju Jang Lee. Tracking moving object using Snake's jump based on image flow[J]. Mechatronics,2001,(11):119-216.
  • 6Szeliski R, Terzopoulos D. Physically-based and probabilistic models for computer vision[A]. Proc SPIE[C]. 1991,1570:140-152.

同被引文献30

引证文献2

二级引证文献19

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部