基于图像处理的车牌识别研究

针对智能交通管理系统中的车牌识别问题,提出应用图像处理技术对汽车的牌照进行识别。车牌定位（LPL,License Plate Location）、车牌分割（LPS,License Plate Segmentation）、车牌识别（LPR,License Plate Recognition）是实现车牌识别系统的最主要的三个部分。先采用HSV模型和RGB模型识别与分割彩色图像,并初步定位车牌图像;再采用radon变换实现车牌的倾斜校正,用投影法对车牌进行定位和分割;最后通过语音读出识别到的车牌信息。通过MATLAB编程进行实验仿真,结果表明利用图像处理技术能够快速地识别出汽车牌照,是一种研究车牌识别的有效方法。

License Plate Recognition Based on Image Processing

The application of image processing technique was put forward to identify the license plate aiming at solving the problem of license plate recognition in the intelligent traffic management system. License plate location, character segmentation, and license plate recognition are the most important three parts, which constitute the license plate recognition system. First, using HSV model and RGB model to identify and segment the color image, and preliminary positioning for the license plate image. Then license plate tilt correction was processed by Radon transform, and using the projection method for license plate location and segmentation. At last, read the result of license information recognized. Simulation by MATLAB programming and experiment,the results show that using image processing technology can quickly identify the license plate, and is an effective method to study the license plate recognition.