摘要
车辆检测是实现视频交通监控技术的核心。在此提出一种基于图像边缘直方图特征的车辆检测方法。为了提高运算的速度,选择较为理想的亮度计算公式将彩色图像转换为灰度图像,然后使用Roberts算子计算图像边缘。通过分析边缘直方图的特征,提出两个特征参数实现车辆的检测。实验表明,该方法有较高的车辆检测准确率,并且能满足智能交通系统中的实时性要求。
Vehicle detection is the key to Video Traffic Monitoring System. A method based on the features of edge histgram is proposed to detect the vehicle. An appropriate formula is chosen to realize the image conversion from color to grey,and then caculate edge of the image using Roberts algrithm. It analyzes the features of edge histogram. Furthermore,two parameters of the histogram is proposed to realize the detection of vehicle. Experiments show that the method has high accuracy rate of vehicle detection and the algorithm can meet the real - time requirements of intelligeiat transportation system.
出处
《现代电子技术》
2009年第14期130-131,134,共3页
Modern Electronics Technique
关键词
车辆检测
检测窗口
边缘检测
边缘直方图
vehicle detection
detection window
edge detection
edge histogram
