基于傅里叶描述子的车牌号码识别

License Plate Number Recognition based on Fourier Descriptor

车牌号码识别基于图像处理、计算机视觉与模式识别技术,在智能交通系统(Intelligent Transportation System,ITS)研究中占有重要的地位.傅里叶描述子是一种常用的描述物体形状和边界的表示方法,它具有对平移、旋转、缩放等几何变化不敏感的特性.使用傅里叶描述子对常见车牌中的字母及数字的边界进行匹配识别,主要分为对车牌的预处理,对字母或数字的边界提取,对边界点的傅里叶描述子计算,与标准模板的匹配过程.实验结果表明利用傅里叶描述子可以有效地识别出存在一定几何失真的车牌号码.

The license plate number recognition based on image processing,computer vision and pattern recognition technology,it plays an important role in intelligent transportation system（ ITS） research. Fourier descriptor is commonly used describing shape of object and boundary representation method,it has the characteristic of translation,rotation,scaling and other characters that geometry not sensitive to the change. Using Fourier descriptors to detect letters and numbers in common license plate boundary matching recognition,the method divided into the license plate pretreatment,letters or numbers of boundary extraction,calculation of the Fourier descriptor of boundary points,and the standard template matching process. The experimental results show that using Fourier descriptors can effectively identify the existence of a license plate number of geometric distortion.