摘要
车辆检测技术是现代智能运输系统的重要组成部分,现有的相关视频检测算法能够检测目标且对环境具有一定的适应性,但其在算法实时性、识别率等方面仍有待提高。提出了一种基于Fisher准则函数法的自适应阈值背景减法和对称差法相结合的运动车辆检测算法,该方法采用surendra算法提取背景,通过背景减法提取出目标前景,再将其与对称差法相结合得到准确的运动目标区域并实时地完成背景更新。实验表明该方法快速、准确,具有一定的实用价值。
Vehicle detection technology is an important part of the modern intelligent transportation systems.The existing video detection algorithms can detect moving targets and have some adaptability to the environment,but the recognition rate and real-time of these algorithms are still to be improved.A novel algorithm for moving vehicle detection,which combined adaptive threshold background subtraction,based on Fisher criterion function method,and symmetrical differencing method,is presented in this paper.This method adopted surendra algorithm to obtain a background image,and then the target foreground is gained using background subtraction,which is combined with the symmetrical differencing to gain more accurate moving target region and update background real-time.The results of experiments proved that the presented algorithm runs rapidly and accurately and has some practical value.
出处
《液晶与显示》
CAS
CSCD
北大核心
2012年第1期108-113,共6页
Chinese Journal of Liquid Crystals and Displays
关键词
车辆检测技术
背景减法
对称差法
背景更新
vehicle detection technology
background subtraction
symmetrical differencing
background updating
