混合模型的运动目标检测与跟踪算法  

Moving object detection and tracking algorithm based on mixture model

作　　者：郎晓彤[1] LANG Xiaotong(Zunyi Normal University,Zunyi 563000,China)

机构地区：[1]遵义师范学院

出　　处：《现代电子技术》2020年第3期70-73,77,共5页Modern Electronics Technique

摘　　要：针对运动目标检测与跟踪算法计算准确率低、效果差的问题,结合Kalman滤波算法,提出一种基于高斯混合模型的运动目标检测与跟踪算法。对传统的权重更新机制进行改进,以保证背景的持续更新状态。建立高斯混合模型,对运动目标图像进行处理,对图像中的参数信息进行实时更新,在Kalman滤波器的性质影响下进行目标跟踪,将获取的运动目标位置作为信息输入,完成运行并跟踪。在更新的过程中,根据运动目标的状态确定矩阵,目标遮挡也不会影响追踪的效果。结果表明,利用Kalman滤波器对运动目标进行跟踪,可以取得良好的运动目标跟踪效果。In combination with Kalman filtering algorithm,a moving object detection and tracking algorithm based on Gaussian mixture model is proposed to deal with the deficiencies of low accuracy and poor effect in conventional moving object detection and tracking algorithm,and improve the traditional weight updating mechanism to ensure continuous updating status of the background.The Gaussian mixture model is created to process the moving object image,update in real time the parameter information in the image and track the object with the performance influence of Kalman filter.The acquired moving object position is taken as information input to complete the operation and tracking.In the process of updating,matrix is determined according to the status of the moving object.In the case of the object occlusion,the tracking effect will not be affected.The results show that the Kalman filter can achieve good tracking effect of moving objects.

关 键 词：高斯混合模型 运动目标检测 跟踪算法 权重更新 目标遮挡 运动目标跟踪 

分 类 号：TN911.1-34[电子电信—通信与信息系统]