改进的PBAS算法在抛物识别中的应用研究

Application of Improved PBAS Algorithm in Throwing Recognition

为了提高在监控视频下进行抛物识别的快速性及准确性,消除ghost区域存在的干扰以及PBAS算法存在的动态背景效果差的问题,提出了一种改进的PBAS算法。介绍了背景减除法与PBAS算法的理论基础,然后提出了一种改进的PBAS算法,并应用于抛物识别中,最后分别在纯净背景和复杂背景下进行抛物识别对比实验。实验证明文中改进的PBAS算法能够完全去除ghost带来的影响,该算法具有更高的检测率和准确率,实现了更好的抛物识别效果,有效提高了识别的平均处理速度,可以在监控视频中更精准、实时地呈现抛物现场的视频数据等信息,更好地满足在抛物识别中的实际应用。

In order to improve the speed and accuracy of throwing recognition in surveillance video,eliminate the interference in ghost region and the poor dynamic background effect of PBAS algorithm,this paper studies and proposes an improved PBAS algorithm.Firstly,the theoretical basis of background subtraction and PBAS algorithm is briefly introduced.Secondly,an improved PBAS algo rithm is proposed by combining the two algorithms.The algorithm is applied to throwing recognition.Finally,the throwing recognition experiments are carried out under two different backgrounds.Experiments show that the improved PBAS algorithm can completely re move the impact of ghost.And the algorithm has higher detection rate and accuracy,and achieves better throwing recognition effect.It effectively improves the average processing speed of recognition,and presents more accurate and real-time information such as throw ing field video data in surveillance video,so as to better meet the practical application in throwing recognition.