基于区域生长法的太极拳竞赛场地分割算法

打开文本图片集

【摘要】本文提出一种基于区域生长的太极拳竞赛场地分割算法。首先，将彩色太极拳竞赛图像转换到YCbCr空间;其次，利用大津法进行阈值分割得到竞赛场地的种子区域;最后，以种子区域作为起始，利用区域生长法分割出太极拳竞赛场地区域。实验结果验证本文算法的有效性。

【关键词】    太极拳竞赛场地分割    区域生长法    种子区域

Taijiquan competition ground segmentation algorithm based on region grow

Sun Rui-yang1， Sun Yu-bin1， Duan Lian2， Zhao Lan-fei3

（1. College of National Traditional Sports， Harbin Sport University， Heilongjiang Harbin 150008;

2. Financial Department， Harbin Sport University， Heilongjiang Harbin 150008;

3. The Higher Educational Key Laboratory for Measuring and Control Technology and Instrumentations of Heilongjiang Province， Harbin University of Science and Technology， Heilongjiang Harbin 150080）

Abstract：  This paper proposes a Taijiquan competition ground segmentation algorithm based on Ransac model. Firstly， taijiquan competition image is transformed into YCbCr color space. Secondly， Otsu is employed to estimate the threshold which segments the competition ground into seed region. Finally， region grow is employed to segment the taijiquan competition which begins from the seed region. Experimental results verify the effect of the proposed algorithm.

Keywords：  taijiquan competition field segmentation; region grow; seed region

引言

隨着科技的不断进步和发展，体育竞赛也逐渐融入了许多科技因素从而使体育运动变得科技化、智能化。（剩余2664字）