计算机科学 ›› 2010, Vol. 37 ›› Issue (3): 271-274.

• 图形图像及体系结构 • 上一篇    下一篇

基于自适应水平集方法的运动目标跟踪

王威,陈益稳,王润生   

  1. (海军航空工程学院青岛分院 青岛266041);(国防科技大学A"I'R重点实验室 长沙410073)
  • 出版日期:2018-12-01 发布日期:2018-12-01
  • 基金资助:
    本文受武器装各预研项目(914008002010705)资助。

Moving Targets Tracking Based on Adaptive Level Set Algorithm

WANG Wei,CHEN Yi-wen,WANG Run-sheng   

  • Online:2018-12-01 Published:2018-12-01

摘要: 在对复杂背景条件下的非刚体目标进行跟踪时,水平集方法是一种非常有效的方法。针对水平集的初始化问题,提出了一种自适应初始化的水平集方法。首先采用粒子滤波对目标进行跟踪,得到运动目标的粗外接矩形轮廓;然后以外接矩形轮廓的质心为基础进行水平集曲线演化,求取目标的精确轮廓信息,并将轮廓的提取结果反馈到跟踪框架中;最后,动态更新参考模板,以改进似然函数。实验结果表明,本方法能够适应非刚体目标在运动中的自由变化,使得跟踪更加精确。

关键词: 粒子滤波,水平集,运动目标跟踪,初始化

Abstract: Level set algorithm is an effectual algorithm when tracking the nonrigid moving targets on complicated condilions. An adaptive algorithm was proposed to solve the initialization problem of level set. Firstly, the particle filter algorithm was used to track the targets, and the rough external contour could be gotten. Then based on the centroid of the external contour,the level set curve evolution was performed to get the precise external contour of the targets,and the extract results were fed back to the tracking frame; lastly the likelihood function was improved by updating the reference template dynamically. hhe experiment results show that the algorithm proposed in this paper can adapt to the moving changes of the nonrigid targets,and the targets can be tracked accurately.

Key words: Particle filter, Level set, Moving targets tracking, Initialization

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!