基于字符行特征的车牌倾斜矫正算法

Approach of Rectifing Distorted Vehicle License Image Based on Its Character Line

针对Hough算法在识别白色反光车牌时的不足,提出了基于车牌字符行特征的倾斜矫正算法.该方法采用通过字符行倾斜度恢复车牌原貌的机理,即从多个角度来对图像倾斜的方向扫描,分别记录各扫描平行线组中具有车牌字符信息的行数,对应字符行数目最多的扫描角度就是图像倾斜的角度.实验表明,该算法能较理想地减少车牌倾斜给车牌字符分割及识别带来的负面影响,运算量小且易行可靠.

An approach of rectifying distorted vehicle license plate which is based on its character line distortion is presented, since Hough-transform does have certain insufficiency in recognizing reflecting white vehicle license plate. The principle is to get the correct license image from its tilt character line. Scanning the tilt image from several different angles, the tilt degree can be calculated by judging the direction that leads to obtain the most character lines. Experimental results show that this method is of great help in reducing those unfavorable influences caused by the tilt of vehicle license plate, which executes a small scale of operation and can be implemented easily.