目前最常用也最有效的运动目标检测方法是背景差法,其中背景提取是背景差法的核心。主要比较了均值滤波法、中值滤波法、混合高斯分布法三种常用的背景建模的方法,说明了各算法在不同情况下的性能优劣。通过实验发现,均值滤波法和中值滤波法适合特定的场合,混合高斯法在场景复杂的情况下效果更好。
In order to obtain accurate real-time traffic information parameters, the use of wireless sensor network technology and GPRS technology as an intelligent traffic-aware front-end and means of communication builds traf- fic information collection network and provides a more convenient solution for traffic information collection. Sensor nodes are designed for the traffic information detection and vehicle monitoring algorithm and the speed measure- ment algorithm are proposed. Experimental results show that the vehicle monitoring algorithm and the speed mea- surement algorithm can be implemented on the sensor nodes with limited capacity and get high accurate real-time traffic parameters, such as traffic flow, speed and so on. Traffic information collection network can meet the needs of large-scale network deployment and also can be used in intelligent traffic system.