License plate recognition system has been widely usedas intelligent traffic system (ITS) the indispensable important constituent part.The license plate recognition algorithm is mature but there are still many places needs to be improved.Based on the principle of license plate recognition, this paperputs forward some new algorithm in the license plate image preprocessing and pattern recognition, includingthe tilt correction processing.Not only reduce the license plate image quality requirements, improve the recognition accuracy, but also improve the recognition accuracy. On the basis of improved algorithm, this paper, by using OpenCV and MFC to complete the algorithm, and using this software to process a series of license plate images has made great identification results.%车牌识别系统作为智能交通管理系统（ITS）中不可缺少的重要组成部分已得到广泛应用。目前的车牌识别算法正在逐步趋于成熟但还是有很多地方有待改进。基于车牌识别的原理，本文在车牌图像预处理以及模式识别两个关键部分提出了一些新的算法，加入了倾斜校正的处理过程，不仅降低了车牌图像的质量要求，而且提高了识别精度。在改进算法的基础上，利用OpenCV结合MFC进行了该算法的软件实现，并利用该软件处理了一系列的车牌图片取得了不错的识别结果。
展开▼
