当前位置: X-MOL 学术Int. J. Syst. Assur. Eng. Manag. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Adaptive strategy for sports video moving target detection and tracking technology based on mean shift algorithm
International Journal of System Assurance Engineering and Management Pub Date : 2021-05-13 , DOI: 10.1007/s13198-021-01128-5
Hongquan Yu , Amit Sharma , Parv Sharma

The continuous improvement in the level of sports competition has led to many recent research designs for providing easy and quick ways for athlete training. The aim behind this research is to present an adaptive hybrid non-rigid target tracking method by adopting (Mean-shift) and color histogram algorithm to process the characteristics of sports video. This work attempts in designing a tracking algorithm by implementing mean shift algorithm for tracking the object characteristics of sports objects. The experimental analysis presents the ideal effects of proposed approach in precision tracking. Mean shift algorithm uses the gradient method to iteratively calculate the extreme points of the probability density function using its characteristics of no parameters and fast pattern matching. In order to realize the tracking of human targets in sports videos, a tracking approach combining the mean shift process and the color histogram process is proposed. Using the statistical robustness of the mean shift process and the characteristics of rapid convergence along the direction of the density gradient, matching of the color histogram to the target shape is done. It solves the problem of variable target shape and high tracking complexity. The proposed method yields 96.04% precision and 97.10% accuracy value for tracking and recognition. The experimental outcomes obtained for the research provides the suitable evidence that the approach presented in this paper has an ideal effect.



中文翻译:

基于均值漂移算法的运动视频运动目标检测与跟踪自适应策略

体育竞赛水平的不断提高导致了许多新近的研究设计,它们为运动员的训练提供了简便快捷的方法。这项研究的目的是提出一种自适应混合非刚性目标跟踪方法,该方法采用(均值漂移)和颜色直方图算法来处理体育视频的特征。这项工作试图通过实现均值平移算法来跟踪运动对象的对象特征来设计跟踪算法。实验分析表明了该方法在精确跟踪中的理想效果。均值平移算法使用其无参数和快速模式匹配的特征,使用梯度法迭代计算概率密度函数的极点。为了实现对运动视频中人类目标的跟踪,提出了一种结合均值漂移过程和颜色直方图过程的跟踪方法。利用平均偏移过程的统计鲁棒性和沿密度梯度方向的快速收敛特性,可以完成颜色直方图与目标形状的匹配。解决了目标形状可变,跟踪复杂度高的问题。所提出的方法产生了96.04%的精度和97.10%的精度值,用于跟踪和识别。为该研究获得的实验结果提供了合适的证据,证明本文提出的方法具有理想的效果。利用平均偏移过程的统计鲁棒性和沿密度梯度方向的快速收敛特性,可以完成颜色直方图与目标形状的匹配。解决了目标形状可变,跟踪复杂度高的问题。所提出的方法产生了96.04%的精度和97.10%的精度值,用于跟踪和识别。为该研究获得的实验结果提供了合适的证据,证明本文提出的方法具有理想的效果。利用平均偏移过程的统计鲁棒性和沿密度梯度方向的快速收敛特性,可以完成颜色直方图与目标形状的匹配。解决了目标形状可变,跟踪复杂度高的问题。所提出的方法产生了96.04%的精度和97.10%的精度值,用于跟踪和识别。为该研究获得的实验结果提供了合适的证据,证明本文提出的方法具有理想的效果。跟踪和识别的精度值为10%。为该研究获得的实验结果提供了合适的证据,证明本文提出的方法具有理想的效果。跟踪和识别的精度值为10%。为该研究获得的实验结果提供了合适的证据,证明本文提出的方法具有理想的效果。

更新日期:2021-05-13
down
wechat
bug