基于小波分解的车辆视频检测算法

作为智能交通系统(ITS)的基础部分,车辆检测系统在ITS中占有很重要的地位,目前常用的基于视频的车辆检测方法主要有:灰度比较法、背景差法、帧差法、边缘检测法.本文在分析这几种方法的优缺点的基础上,提出了一种基于数学形态学滤波和小波分解的算法.该算法首先对视频图像进行形态学滤波,然后在虚拟检测区进行小波分解,通过分析小波系数来检测车辆,它计算量小,复杂度低,可正确判断有无车辆、完成车辆的计数,实现车流量计算.

Algorithm of Vehicle Detection by Video Image Based on Wavelet Decomposition

As a basic part of intelligent transportation system (ITS),vehicle detection system plays a important role in the ITS.Currently,there are several detection ways by video,such as gray comparison,background difference,frame difference and edge detection.This paper is on the base of analysis advantage and disadvantage of these algorithms,a new algorithm is proposed based on the mathematical morphology filter and wavelet decomposition.This algorithm uses gray morphology filter firstly,then makes wavelet decomposition on the virtual detecti on area,lastly,detects vehicle by analyzing wavelet coefficient variance.This algorithm computational complexity is lower than others,which can judge if the vehicle is passed,count vehicles passed virtual area,accomplish vehicle flow computation.