Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (17): 148-150.DOI: 10.3778/j.issn.1002-8331.2009.17.045

• 图形、图像、模式识别 • Previous Articles     Next Articles

New method for license plate location and application

LI Yu-cheng,QIN Ni,WANG Mu-shu   

  1. Department of Automation,North China University of Technology,Beijing 100144,China
  • Received:2008-09-24 Revised:2008-12-19 Online:2009-06-11 Published:2009-06-11
  • Contact: LI Yu-cheng

一种新的车牌定位方法及应用

李宇成,覃 妮,王目树   

  1. 北方工业大学 自动化系,北京100144
  • 通讯作者: 李宇成

Abstract: For the limitation of current license plate location measurement methods,a new algorithm is presented,using complicated texture,sharp-cut contrast and exterior to get the contrast energy,based on extremum region of convolution energy.It also has the advantage to avoid the difficulty to choose a threshold between 0~255 compared with other algorithms.Large numbers of experiments show that this algorithm have higher accuracy and robust.Finally explain how to apply this method to carry on the vehicle speed measurement.

摘要: 针对现有车牌定位方法的局限性,提出了一种基于灰度图像求取卷积能量极值区域的车牌定位新方法。该方法充分利用车牌的纹理复杂、对比度鲜明、外型规则等特征构造车辆图像的能量图,然后通过选取能量区域定位车牌。其明显优点是,避免了常见方法中直接对图像进行二值化时阈值难以确定的问题。大量实验表明该方法计算简单,准确率高,鲁棒性好。最后说明了如何应用该方法进行车速测量。