东北大学学报（自然科学版）

东北大学学报(自然科学版) ›› 2011, Vol. 32 ›› Issue (11): 1558-1561.DOI: -

• 论著 • 上一篇    下一篇

视频运动目标提取的实现

闫爱云;李海朋;李晶皎;王骄;   

  1. 东北大学信息科学与工程学院;
  • 收稿日期:2013-06-19 修回日期:2013-06-19 发布日期:2013-04-04
  • 通讯作者: -
  • 作者简介:-
  • 基金资助:
    国家自然科学基金资助项目(60970157)

Implementation of moving object extraction in video

Yan, Ai-Yun (1); Li, Hai-Peng (1); Li, Jing-Jiao (1); Wang, Jiao (1)   

  1. (1) School of Information Science and Engineering, Northeastern University, Shenyang 110819, China
  • Received:2013-06-19 Revised:2013-06-19 Published:2013-04-04
  • Contact: Yan, A.-Y.
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

718

摘要： 针对混合高斯模型建模实现背景差分获取的运动目标中含有较多的阴影这一问题,对基于Phong光照模型的判断和检测阴影的准则作了改进,并给出了视频运动目标检测算法.使用基于混合高斯模型的背景差分法和对称差分法相结合获取运动目标,使用改进的阴影判断准则检测和去除阴影.结果表明:改进后的阴影判断准则去除阴影所用时间比原方法平均减少了18%~30%,并且能够取得较好的去除阴影效果,最终获得精确的运动目标.

关键词: 目标检测, 背景差分, 对称差分, 去阴影

Abstract: There are many shadows in moving target detection area which obtained by background subtraction based on Gaussian mixture model. Criterion of detecting shadow and judgment of Phong illumination model was improved, and a video moving object detection algorithm was presented. Combining the background subtraction and symmetry subtraction on Gaussian mixture model to get moving object, and using improved algorithm to remove shadow. The results showed that the improved standards reduce the average deal time by 18% to 30% and still obtain better results of removal shadow, and finally gain the precise motion target.

中图分类号: