提出了一种先分段，再利用Hough变换拟合直线的水平分割方法和基于先验知识约束的垂直分割方法，同时提出目标增强的预处理方法．大量实验表明，该算法较好地解决了以上问题，适用于各种质量的车牌图像，分割的准确率较高．-Presents a first section, then use the level Hough transform fitting straight line segmentation method and prior knowledge constraints on vertical segmentation method, while enhancing the preprocessing method proposed objectives. A large number of experiments show that the algorithm can solve the above problems, for various quality license plate image, the higher the accuracy of segmentation.