摘要：

针对道路车辆的实时监测系统检测效率和精度较低的问题,设计一套实时监测系统。对视频图形进行灰度化、滤波及增强,并分割出统计区域。在分割出的图像中,设计统计众值法构建背景模型,设置阈值获得前景图像,采用Canny算子检测车辆边缘,将前景图像与车辆边缘叠加进行形态学运算以获取车流量统计结果。实验结果表明,该系统准确率高达98.45%,能够满足智能交通系统对检测效率和精度的需求。

关键词: 智能交通, 背景建模, 统计众值法, 帧间差分, 车辆统计

Abstract:

To solve the problem that the detection efficiency and accuracy of the real-time monitoring system for vehicle are low,a real-time vehicle detection system is designed.First,the system grayscales,filters,and enhances the video images to segment the statistical regions.Then,in the segmentation image,the statistical mode method is designed to construct the background model and the Canny operator is used to detect the vehicle edge.At last,the foreground and vehicle edge images are superimposed for morphological operation to obtain the vehicle counting.Experimental results show that the accuracy of the system is up to 98.45%,which can meet the needs of intelligent transportation systems for detection efficiency and accuracy.

Key words: intelligent transportation, background modeling, statistical mode method, frame difference, vehicle counting

中图分类号: 

• TP277