Novel License Plate Image Preprocessing Approach for Character Segmentation
一种新的面向字符分割的车牌图像预处理方法

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.