我国经济的快速发展使得汽车拥有量急速增加，公路交通逐渐成为比较重要的运输途径，所以公路交通是我国大力发展的基础设施。城市交通的日益拥堵需要用更有效和更先进的交通管理和控制。用电子信息技术提高交通效率、管理效率以及安全,智能交通系统 ITS 已成了目前交通管理的主要方向。 本文将现代信号处理技术——小波技术，应用到智能交通领域，即将小波变换应用到车牌识别技术当中来，并结合改进的BP神经网络和模板匹配对车牌字符进行分类识别。主要的研究内容和成果如下： 1、在预处理阶段，本文使用了灰度变换、邻域滤波、中值滤波进行图像降噪，使用直方图均衡化和灰度拉伸进行图像增强，为了突出车牌信息，在直方图均衡化的基础上，本文提出了基于小波变换的非线性图像增强算法。经试验证明使用该算法平滑了图像的直方图，而且在整体上增强了图像的清晰程度，并且图像的边缘有了明显的增强。 2、在车牌定位阶段，本文使用了dbN小波的车牌定位算法。通过研究dbN小波的特点和性质，并通过大量的仿真实验，提出了采用db8小波对车牌的左右边界进行定位。在采用数学形态学等方法取得车牌的上下边界以后，使用db8小波对图像进行分解处理，提取有效信息。经试验证明，db8小波能够有效的去除车灯等无关区域的影响，为最后的车牌细定位提供良好的支持，最终本文算法对车牌的定位准确率达到91.2%。 3、在字符分割阶段，把车牌图像首先进行了二值化和倾斜角的校正处理。通过大量的实验，提出一种基于小波变换了的阈值函数，对车牌区域进行消噪；然后是字符分割，先除去车牌上下边框，利用垂直投影、先验知识结合的办法分割出每一个字符区域，再寻找字符的形态学连通域，切出字符矩形最小区域，对误切分汉字进行合并处理。 4、对于字符的识别，先对字符图像进行预处理，包括大小归一化和细化处理；然后研究了字符的特征，针对BP神经网络收敛速度慢，在汉字字符识别的过程中，提出了基于改进的的BP神经网络进行识别；最后提出了一种改进的模板匹配算法，将模板匹配与字符特征和提取边缘模板结合在一起，来识别数字和字母。此方法使得正确率和适应性都得到了提高，具有很好的应用价值。

With the rapid development in China’s economic make the car ownership increased,highway traffic has become a major transport pathway and it becomes the infrastructure to develop of the country. The more efficient traffic management is needed because of the growth of traffic.Intelligent Transportation System(ITS) has become the main direction of the current traffic management system, which is more efficienct in management ,transport and safety. In this paper, modern signal processing techniques-wavelet technology is applied to the Intelligent Transportation System license plate recognition technology, which combined with the improved BP neural network and template matching for classification of the license plate characters.The main research contents and results are as follows. 1, In image preprocessing phase, the paper uses the gray-scaled transformation, median filter and neighborhood filter for image noise reduction, uses histogram equalization and gray stretch for image enhancement. In order to highlight the license plate information, this paper uses a new wavelet transform algorithm. It is proved that the enhancement algorithm using wavelet transform can smooth the histogram and overall the image bright, and the edge has been significantly enhanced. 2, In processing of license plate location, this paper uses the license plate location algorithm dbN wavelet. By studying the characteristics and nature of dbN wavelet, and by the large number of simulation experiments, the db8 wavelet on the left and right borders of the license plate positioning. Db8 wavelet decomposition processing the image using mathematical morphology and other methods to obtain the license plate of the upper and lower boundary, and extract useful information. The test proved that the db8 wavelet can effectively remove the lights and unrelated to the region,it provide good support for the license plate of fine positioning, the final positioning accuracy of the algorithm on the license plate up to 91.2%. 3, The character segmentation stage, firstly make the license plate image binarization and the correction of the tilt angle. By a large number of experiments, the threshold function based on the wavelet transform to eliminate the noise of the plate area; and character segmentation, remove the license plate up and down the border, use the vertical projection of a priori knowledge and the way split of each a character area, and then look for the character of morphological connected domains, and cut out the character rectangle of minimum area, merge the mistakenly segmentation Chinese character. 4, For character recognition, character of image must be preprocessed, including size normalization and refinement process; then study the characteristics of the characters and the slow rate of convergence for BP neural network in the process of Chinese character recognition,based on the improvement of the BP neural network to identify; an improved template matching algorithm be put, template matching combined with character features and extract the edge of the template to identify the numbers and letters. This method allows the correct rate and adaptability have been improved, with a good application value.