摘要
车辆定位在车辆跟踪中有着至关重要的作用。为了更加精确地定位运动车辆,提出一种基于多尺度的小波变换方法定位目标奇异点。在该方法中,先后采用较大尺度的Marr小波以及小尺度Gauss函数一阶导数小波变换实现对车辆宽度的定位,分别采用大尺度Marr小波以及小尺度Marr小波实现对车辆长度的定位,根据宽度定位中的最复杂点与长度定位中最复杂点的交集实现对车辆的精准定位。实验结果表明,该方法能较好识别运动车辆,实现对车辆的精准定位。
Vehicle positioning plays a vital role in vehicle tracking.In order to locate the moving vehicles more accurately,a multi-scale wavelet transform method is proposed to locate the target singularities.In this method,we use Marr wavelet in large scale and small scale Gauss function derivative wavelet transform to position the vehicle width,respectively realize the positioning of vehicles length with large scale and small scale Marr wavelet to achieve precise positioning of the intersection of the vehicle through the most complex positioning in width and length the most complex point positioning.The experimental results show that the method can identify the moving vehicles well and realize the accurate positioning of the vehicles.
作者
黄恒
管庶安
Huang Heng;Guan Shu’an(Wuhan Polytechnic University,Wuhan Hubei 430040,China)
出处
《信息与电脑》
2018年第2期33-34,共2页
Information & Computer
关键词
小波变换
宽度定位
长度定位
wavelet transform
width location
length location