摘要
为有效提升交通管制水平,研究基于光学相控阵的交通视频图像运动目标检测方法。通过由光学相控阵所产生的可控激光网搜索到交通视频图像运动目标,将此目标照亮并呈现其成像,以此为依据分析所搜索目标的图像灰度特征,依据此特征通过背景差分法以差分运算的方式获取交通视频图像运动目标的差分灰度图像,采用最大类间方差法对此差分灰度图像实行阈值分割,并通过形态学滤波的腐蚀、膨胀、开闭运算除掉分割后图像的过暗与过亮区信号,降低所得图像的噪声干扰,得到高精度的运动目标图像,实现交通视频图像运动目标检测。以某地区交管部门的交通监控视频图像为实验对象,结果表明,该方法可实现交通视频图像运动目标的有效检测,且检测结果稳定,效果优越,所得结果的图像质量高,具有非常高的实际应用价值。
In order to effectively improve the level of traffic control,a method of traffic video image moving target detection based on phased optical array is studied.By phased optical array produced by controlled laser web search to traffic video image moving targets,the target light and render its imaging,based on the search target image grey feature analysis.Based on the features obtained employing background difference method with the differential operation of traffic video image moving targets difference grey image,by adopting the method of between-cluster variance to this difference grey image threshold segmentation.Through the morphological filtering of corrosion,expansion,after the opening and closing operation to remove segmentation image too dark and light signal,and reduce the noise of the image,get high accuracy of the movement target image,Realisation of traffic video image moving target detection.Taking the traffic surveillance video image of a traffic control department in a particular area as the experimental object,the results show that the method can effectively detect the moving target of traffic video image;the detection result is stable,the effect is superior,the resulting image is of high quality and has very high practical application value.
作者
代艳萍
郑彬
张健
DAI Yanping;ZHENG Bin;ZHANG Jiang(College of Transportation and Automobile Engineering,Panzhihua University,Panzhihua Sichuan 617000,China)
出处
《激光杂志》
CAS
北大核心
2021年第7期45-49,共5页
Laser Journal
基金
四川省教育厅自然科学重点项目(No.18ZA0293)
国家自然科学联合基金项目(No.U1960101)。
关键词
光学相控阵
交通视频
运动目标
激光网
背景差分
灰度特征
phased optical array
traffic video
moving target
laser net
background difference
grayscale characteristics
