东北大学学报:自然科学版 ›› 2016, Vol. 37 ›› Issue (6): 770-774.DOI: 10.12068/j.issn.1005-3026.2016.06.003

• 信息与控制 • 上一篇    下一篇

基于光流法的运动目标检测与跟踪算法

肖军1, 朱世鹏2, 黄杭2, 谢亚男3   

  1. (1. 东北大学 信息科学与工程学院, 辽宁 沈阳110819; 2. 东北大学 计算机科学与工程学院, 辽宁 沈阳110819;3.北京理工大学 计算机学院, 北京 100081)
  • 收稿日期:2015-05-18 修回日期:2015-05-18 出版日期:2016-06-15 发布日期:2016-06-08
  • 通讯作者: 肖军
  • 作者简介:肖军(1967-),女,辽宁沈阳人,东北大学教授.
  • 基金资助:
    国家自然科学基金资助项目(61201054).

Object Detecting and Tracking Algorithm Based on Optical Flow

XIAO Jun1, ZHU Shi-peng2, HUANG Hang2, XIE Ya-nan3   

  1. 1. School of Information Science & Engineering, Northeastern University, Shenyang 110819, China; 2. School of Computer Science & Engineering, Northeastern University, Shenyang 110819, China; 3. School of Computer Science, Beijing Institute of Technology, Beijing 100081, China.
  • Received:2015-05-18 Revised:2015-05-18 Online:2016-06-15 Published:2016-06-08
  • Contact: XIAO Jun
  • About author:-
  • Supported by:
    -

摘要: 选用Harris角点作为跟踪对象,将尺度空间引入角点检测,提取特征尺度上的Harris角点,并进行曲率非极大值抑制,滤除“伪角点”,提高角点检测对尺度变化的抗扰能力.跟踪算法选用结合图像金字塔的光流法,迭代计算光流,并提出基于光流误差的跟踪算法,即用不同时间流的运动轨迹在同一帧图像的误差来衡量运动跟踪情况,避免跟踪点因被遮挡、消失或者纹理特征发生变化而导致跟踪失败.通过对不同视频图像进行检测的结果证明基于改进的角点提取和图像金字塔的光流法具有良好的跟踪效果,引入光流误差可以有效地滤除跟踪失败的特征点,准确估计运动目标的位置.

关键词: 目标跟踪, 角点, 特征尺度, 光流法, 图像金字塔

Abstract: Harris corner points were adopted as tracking objects, and scale space was introduced into corner point detection in order to extract Harris corner points in feature scale. Then curvature was computed to filter out false corners and enhance adaptability to scale change. Optical flow method was adopted for the tracking algorithm based on image pyramid, in which the optical flow iteratively was computed. And the tracking algorithm based on the optical flow error was proposed. That is, the trajectory error in the same frame with different time flow was used to evaluate the tracking situation. In this way, tracking failure was avoided when the tracking object is hidden, disappears or textural features change. Experimental results of different video sequences show that the proposed optical flow tracking algorithm based on improved corner extraction and image pyramid has better tracking performances. The feature points could be filtered effectively that lead to tracking failure with the introduction of optical flow error method, and the object positions are estimated accurately.

Key words: object tracking, corner point, feature scale, optical flow, image pyramid

中图分类号: