一种新的面向字符分割的车牌图像预处理方法

Novel License Plate Image Preprocessing Approach for Character Segmentation

首先进行图像增强以滤除噪声、校正不均匀光照 ,增强对比度 ,然后分析了车牌的倾斜模式 ,提出了一种新的车牌倾斜校正方法。该方法采用水平和垂直边缘检测结合旋转投影求取车牌的水平和垂直倾斜角度并通过垂直边缘检测进行上下位置精定位 ,对水平倾斜的车牌进行双线性插值旋转校正 ,对垂直倾斜的车牌进行双线性插值错位偏移校正。实验结果表明 ,该方法简单实用 ,对光照不均、低对比度。

Firstly,some image enhancement technologies are applied to remove noise,reduce the effects of non-uniform illumination and enhance contrast;then tilt models of a plate are analyzed and a novel algorithm for number plate tilt correction is advanced.Horizontal and vertical incline angle is obtained by combining horizontal and vertical edge of a plate with its rotation projection,respectively,and a refined top and bottom localization of the plate is obtained by its vertical edge.Bilinear interpolation rotation correction is performed to the plate that is tilt to a horizontal line and bilinear interpolation correction on location excursion of pixels in a low is executed to the plate that is tilt to a vertical line.Experimental results show proposed method can be implemented easily and offers robustness when dealing with the number plate that has dirty,noise,poor contrast,non-uniform illumination or gradient.