【摘要】 车牌定位和字符分割是车牌识别系统中关键的部分,定位的准确程度与分割准确性直接影响到车牌识别的效果。基于图像处理技术,对车牌灰度图像的进行灰度修正,Wiener滤波并进行二值化,使之能够准确地从图像中提取出车牌。采用线性映射的方法,增强图像灰度范围,基于水平扫描与先验知识约束条件相结合的算法进行车牌定位,同时利用最小二乘法原理,实现了最大类间方差法对车牌进行分割。用MATLAB软件进行仿真实现,验证效果好。更多还原

【Abstract】 The plate location and character segmentation is a key part of the license plate recognition systems,the positioning accuracy and the segmentation accuracy directly affects the effect of the license plate recognition.Based on the image processing technology,it firstly correct grayscale images,Wiener filtering,binarization in order to extracted the plate from the image the license plate image accurately,Linear mapping is used to enhance image grayscale range,plate location can be obtained based on the horizontal scanning lines combined with prior constraints algorithm.license character segmentation problem can be solved by otsu method based on the least squares method.Simulation experimental with Matlab results proved that the algorithm proposed has advantages of fast license plate localization and character segmentation.更多还原