|Table of Contents|

Efficient Goal-mouth Detection Method Based on SVM

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

Issue:
2010年01期
Page:
13-18
Research Field:
Publishing date:

Info

Title:
Efficient Goal-mouth Detection Method Based on SVM
Author(s):
ZHANG Yu-zhenHE XinWANG Jian-yuDAI Yue-weiFAN Bai-chao
School of Automation,NUST,Nanjing 210094,China
Keywords:
goal-mouth detection support vector machine video retrieval Top-Hat transform
PACS:
TP391.41
DOI:
-
Abstract:
To efficiently detect a goal-mouth with complicated backgrounds in soccer video,this paper proposes a goal-mouth detection algorithm based on support vector machine(SVM).For each frame,a Top-Hat transform is used to enhance white color and the achieved RGB image is converted to a grayscale intensity image and then converted to a binary image.Based on the morphologic connection analysis,the two longest vertical lines meeting some conditions are achieved and seen as potential goalposts and the feature vector can be computed.With the help of strong study ability from SVM,the goal-mouth is detected.Experiments prove that this algorithm is not only efficient,but also has higher robustness and is more flexible compared with the existing algorithms for goal-mouth detection.

References:

[ 1] ?? YangY, Lin S X, Zhang Y D, et a.l H igh lights extrac??tion in soccer videos based on goa l??mouth detection[ A].Proc o f the 9th Internationa l Symposium on S igna l Pro??cessing and Its Applications[ C]. Shar jah, 2007: 1- 4.
[ 2] Yow D, Yeo B L, YeungM, et a ls. Analysis and pres??entation of soccer high lights from dig ita l v ideo [ A ].Proc of ACCV[ C]. [ s. L. ]: [ s. n. ], 1995.
[ 3] Wan KW, Yan X, Yu X G, et a.l Rea l??tmi e goal??mouthdetection in MPEG soccer v ideo [ A ]. Proc o f the 11thACM International Conference on M ultmi ed ia [ C ].Berkeley, Ca lifornia, USA: [ s. n. ], 2003: 311- 314.
[ 4] 庄越挺, 刘骏伟, 吴飞, 等. 基于支持向量机的视频字幕自动定位与提取[ J]. 计算机辅助设计与图形学学报, 2002, 14( 8) : 750- 771.
[ 5] 刘国翌, 杜威, 李华. 足球场地标志线的自动提取[ J]. 计算机辅助设计与图形学学报, 2003,15( 7): 870- 874.
[ 6] 蔡群, 陆松年, 杨树堂. 基于音视特征的视频内容检测方法[ J]. 计算机工程, 2007, 33( 22): 240- 242.
[ 7] Cover T M. Geom etr ica l and statistical properties o fsystem s and linea r inequa lities w ith app lications inpattern recogn ition [ J]. IEEE Transac tions on E lec??tronic Com pute rs, 1965, 14( 3): 326- 334.
[ 8] Cortes C, Vapn ik V. Support v ec tor netw orks [ J].M achine Learning, 1995, 20( 3): 273- 297.
[ 9] Vapnik V. The nature o f sta tistica l learning theory[M ].New Yo rk: Springer, 1995.
[ 10] Courant R, H ilbert D. M ethods o fm athem atical phys??ics, Vo lum e 1[M ]. New York: Intersc ience Publish??e rs Inc, 1953.
[ 11] 周艺华, 曹元大, 张龙飞. 一种新的基于对象的足球视频镜头分类方案[ J] . 计算机工程与应用,2005, 34: 229- 232.
[ 12] 闫龙川, 杜军. 基于SVM 的足球场地禁区检测[ J].计算机应用, 2007( 12): 80- 90.

Memo

Memo:
-
Last Update: 2012-11-02