基于颜色相似度的车牌定位算法

针对车牌图像分割困难、车牌位置定位不准确等问题，提出了一种新的车牌快速定位方法。该方法是在HSV模式下进行的，首先在颜色空间下，对车身的每一行颜色进行相似度的计算，再将相似度大于阈值的颜色去除，然后进行水平与垂直投影，同时利用车牌位置、长宽比例信息，精确地定位出车牌。这种方法对不同光照条件下，不同颜色的车辆，不同颜色的车牌均具有良好的适应性。实验表明该方法定位准确，运算速度快，能满足实时性要求。

License Plate Location Arithmetic Based on Color Similarity

Aimed at the difficulty of license plate location, this paper presents an effective license plate location algorithm based on HSV color model, which employs projection algorithm .At first,the algorithm calculates the similarity of each row in the color image and wipes off the pels whose similarity exceeds threshold. Then project the result in horizontal and vertical direction .Meanwhile, make full use of the information about the position and proportion of license plate. At last ,we can divide license plate correctly. The algorithm has fewer limits to the change of light, the color of car and license plate. So it can be employed to the applied vehicle license plate recognition system. The results show that this algorithm combines the advantages of veracity and of runtime, and fits for real time detection.