基于视频的车辆检测和牌照识别  

Detection of Vehicle and Recognition of License Plate Based on Video

作　　者：李金屏[1,2] 王群[1] 吴翔[1,2] 黄伟[1] 安冬[1] 卢迪迪[1] 郭田梅 范琳伟[1] 王结[1] 

机构地区：[1]济南大学信息科学与工程学院,山东济南250022 [2]济南大学山东省网络环境智能计算技术重点实验室,山东济南250022

出　　处：《济南大学学报（自然科学版）》2014年第2期89-96,共8页Journal of University of Jinan(Science and Technology)

基　　金：国家级大学生创新创业训练项目(201210427017);国家自然科学基金(61203341);山东省高等学校科技计划(J12LN19)

摘　　要：为了提高执法效率和降低人力成本,研究一种基于视频的实时高效监控方案。该方案首先利用混合高斯模型和3σ原则的背景差分法,从视频中自动检测和定位车辆,对其行为轨迹进行标记,然后利用能量等信息获得车辆的牌照信息。实验表明,该系统能够对进入监控区域车辆的轨迹进行准确标定。该方案有助于提高交通视频监控系统的安全性、实时性及可靠性。In order to improve the efficiency of executing the law and reduce the cost of human resources, an efficient and effective scheme is proposed, in which Gaussian mixture model and background differencing method based on 3σ principle are employed. By analyzing the video shot, the vehicle can be located and the trajectory of the vehicle can be also marked;the information of the corresponding license plate can be obtained by using the energy information. Experiments showed that the vehicle in monitored area can be analyzed through tracking, and the security and effectiveness of the proposed scheme can be demonstrated.

关 键 词：智能交通 视频监控 车牌识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]