基于智能手持设备的车牌识别系统研究

Research of Vehicle License Plate Recognition System Based on Smart Handheld Device

摘要:

以交警能更加快捷方便地稽查运营车辆为目的,借助最新的移动互联网技术,分析并研究了多种计算机视觉算法;提出了一种基于K均值聚类(K-means)的车牌定位技术,通过颜色的聚类分析来确定车牌区域,定位车牌;又应用垂直投影技术和模板匹配法来分割和识别车牌字符,利用SQlite开发了具有存储车牌字符信息功能的数据库;在此基础上,研制了一套基于智能手持设备的车牌识别系统;功能测试表明,该系统具有良好的性能,能够较好地识别出车牌信息。

Abstract:

With the latest mobile Internet technology, in order to make it more convenient and easy for the traffic police to check the operation of the vehicles, a variety of computer vision algorithms are analyzed and studied. A license plate location technology based on K-means clustering is proposed. License plate area can be certain and located by color clustering analysis. Then, the license plate characters are segmented and recognized by the vertical projection technique and template matching method. The database which can store the character information of the license plate is developed by using SQlite. On this basis, a vehicle license plate recognition system based on smart hand-held device is developed. The system, which has good stability, can recognize the license plates through functional tests.