基于图像处理的车牌识别技术

汽车车牌照的识别是智能交通系统必不可少的组成部分，也是保障车辆安全的必要手段。本文针对智慧小区设计一种简易实用的车牌识别系统，设计内容包括图像采集及预处理，车牌定位，边缘检测，字符识别等。系统开发构建在Qt软件平台上，对采集的图像进行灰度化、边缘检测二值化处理，同时，通过水平投影进行区域定位消除边框，再利用处置投影对车牌字符分割，最后根据特征值对系统字符识别，输出字符。实验验证，该系统检测方便易行，识别效果较好。

A Recognition Technique for License Plate based on Image Processing

Vehicle license plate recognition technology is an indispensable component of intelligent transport system and necessary means to ensure vehicle safety. A simple and practical license plate recognition system for intelligent residential area is designed in this paper, which include image acquisition and preprocessing, license plate location, edge detection, character recognition and so on. The system is developed and build on Qt software plate, and the grayscale and binarization，edge detection of the collected image are processed. In the meantime, the region is located by horizontal projection to eliminate the border, and then the license plate character is segmented by disposal projection. Finally, the characters are recognized and output according to the characteristic value. Experimental results show that the system is easy to detect and effective.