计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (2): 227-227.

• 工程与应用 • 上一篇    下一篇

足球视频中的目标检测与跟踪

娄娜,何南忠,施保昌   

  1. 武汉市华中科技大学数学系
  • 收稿日期:2006-01-24 修回日期:1900-01-01 出版日期:2007-01-11 发布日期:2007-01-11
  • 通讯作者: 娄娜 louna

Detection and Tracking in Soccer Video Sequences

,,   

  1. 武汉市华中科技大学数学系
  • Received:2006-01-24 Revised:1900-01-01 Online:2007-01-11 Published:2007-01-11

摘要: 本文提出了一种新颖的目标检测与跟踪算法来检测和跟踪足球运动中的球员。与普通的利用颜色分割的方法不同,考虑到足球视频中非目标的像素大体上都是单一的绿色这个特点,结合颜色的统计信息和像素的边缘特性来得到更完美的检测效果;接着利用灰度图像中的统计信息,轻松地完成球员队属的辨别;最后根据重叠面积提出一种简单的视频目标跟踪方案,结合图像匹配,解决运动中的遮挡问题。

关键词: 检测, 跟踪, 颜色统计信息, 边缘特性, 图像匹配

Abstract: This paper propose a novel object detection and tracking method in order to detect and track players in Soccer Video Sequences. Different from other common segment method, the proposed method refers to color rarity and local edge proper considering the characteristics of soccer video images, that most non-object regions are roughly single green. Next identify the team that a player belongs to easily according to statistics of gray in gray images. Last, integrate image match, then we also propose a simple tracking method according to the area, that could track players with occlusion with others.

Key words: detection, tracking, color rarity, local edge, image match