不均匀光照下的灰度车牌图像二值化方法设计与实现

不均匀光照下的灰度车牌图像二值化方法设计与实现Gray License Plate Image Binarization Method under Non-uniform Illumination Design and Implementation＊ 摘要数字图像处理是用一种特定的计算方法对数字图像进行处理的一种过程，在讨论过程中，不均匀光照下的图像二值化是数字图像处理中的一个难题，在汽车牌照自动识别系统工作当中，往往会在特别的场景下遇到复杂的光照环境,因此常常会出现车牌光照不均的现象，给图像二值化带来困难.本文介绍了在不均匀光照下对灰度车牌二值化的一种局部阈值的方法：Bernsen 法。实验表明,在复杂光照下对于车牌的二值化采纳传统的全局阈值法效果并不乐观，而使用局部阈值法能够有效地克服不均匀光照对车牌识别的影响,二值化效果较好，车牌识别率得到显著的提高。关键词：不均匀光照；二值化；局部阈值法Abstract Digital image processing is to use a specific calculation method for digital image processing process of a kind of image processing in the process of research ， the non-uniform illumination image binarization is a difficult problem in digital image processing, the car license plate automatic recognition system of work, often encountered in special situations of complex light environment， so often can appear the phenomenon of uneven illumination ， the license plate bring difficulties.The grayscale plates under non—uniform illumination of a local threshold binarization methods: bernsen method in this paper 。 And comparing the global threshold method ， the experimental results show that under complex illumination for the license plate of the global threshold method with traditional binarization effect is not optimistic， and using local threshold value method can effectively overcome the effects of non—uniform illumination on the license plate recognition ， the effect of binarization ， license plate recognition rate are improved significantly。 Keywords： Non—uniform illumination； Binarization； Loca...