一种车牌图像校正新方法

【摘要】因摄像机角度而造成的机动车牌图像倾斜会对其后继的字符分割与识别带来不利的影响。本文在分析了车牌倾斜模式的基础上,提出了一种基于最小二乘支持向量机（LS-SVM）的车牌图像倾斜校正新方法。通过LS-SVM线性回归算法求取坐标变换矩阵并对畸变图像进行旋转校正。主要方法：首先,将二值倾斜车牌图像中的像素转换为二维坐标样本,并构造图像数据集 再通过LS-SVM线性回归算法对该数据集进行回归,求取主要参数 最后,再由该参数转换为能反映图像倾斜方向的2维坐标变换矩阵。实验结果表明,该方法简便实用,对光照、污迹等不敏感,抗干扰能力强。-New Method of a license plate image correction

Abstract caused due to camera angles, the image tilt motor vehicle license will have on its subsequent recognition of characters segmentation and adverse effects. Based on the analysis of the inclined plate model based on proposed based on least squares support vector machine (LS-SVM) of the license plate Skew new approach. By LS-SVM linear regression algorithm to strike a coordinate transformation matrix and rotate the image distortion correction. Main methods: First, the value of the two inclined plate in the image pixel is converted to two-dimensional coordinates of the sample, and construct the image data sets　then the linear regression through the LS-SVM regression algorithm for the data set to strike a key parameter　final , and then by the parameter is converted to reflect the tilt direction of two-dimensional image coordinate transformation matrix. The experimental results show that the method is simple and practical, to light,