摘要
针对网球比赛中运动目标的有效快速跟踪问题,提出基于红外光谱仪的网球运动目标跟踪方法,采用红外光谱仪中采集模块采集运动目标的干涉图复原光谱,与实际光谱进行相位差比对,构建动态观测模型,并根据观测结果采用局部背景加权标记运动目标,最后通过局部搜索算法自适应更新动态观测模型,完成对运动目标的自动跟踪.实验结果表明,提出的方法在不同的背景、光照、角度变化的情况下能保持较高的跟踪成功率.
In view of the problem of the automatic tracking of tennis and players in tennis competitions,a method of tracking tennis moving targets based on infrared spectrometer is proposed.The interferogram of acquisition module in infrared spectrometer is used to recover the spectrum,the phase difference is compared with the actual spectrum,the dynamic observation model is constructed,and the observation knot is based on the observation junction.The target markers are determined by local background weighting.When the target is detected,the local search algorithm is used to dynamically update the dynamic observation model of the particle to realize the tracking of the moving target.Experimental results show that the proposed method maintains high tracking success rate under different background,illumination and angle changes.
作者
王慧玲
许宁
杨景元
雷魏
WANG Hui-ling;XU Ning;YANG Jing- yuan;LEI Wei(Education College,Tibet University,Lhasa 850000;Institute of Information Science and Technology,Tibet University,Lhasa 850000 China)
出处
《湘潭大学自然科学学报》
CAS
2018年第2期46-49,共4页
Natural Science Journal of Xiangtan University
基金
西藏自治区哲学社会科学专项基金项目(17BTY001)
关键词
红外光谱仪
网球
背景加权
局部搜索
目标跟踪
infrarecl spectrometer
tennis
background weighting
local search
target tracking