Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (22): 199-202.DOI: 10.3778/j.issn.1002-8331.2010.22.058

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

Weaken crest and fill trough adaptive algorithm about license plate location

GUAN Ming-shan,REN Hong-e,WANG Yang,HAN Na   

  1. Department of Computer and Information Engineering,Heilongjiang Institute of Science and Technology,Harbin 150027,China
  • Received:2009-02-25 Revised:2009-04-27 Online:2010-08-01 Published:2010-08-01
  • Contact: GUAN Ming-shan

汽车牌照定位的削峰填谷自适应算法

关明山,任洪娥,王 洋,韩 娜   

  1. 黑龙江科技学院 计算机与信息工程学院,哈尔滨 150027
  • 通讯作者: 关明山

Abstract: A weaken crest and fill trough algorithm used for license plate location is put forward.At first,this algorithm deals the obtained color image with gray,gray stretch and edge detection,so makes the area of license plate sticking out.Then the detected result is projected vertically,and deals the projection with weaken crest and fill trough adaptive targeted,and can get the left-right border of the plate,the part image between the plate’s left and right border is projected horizontally,and deal the projection with same algorithm,and also can get the top-bottom border of the plate.In the end,the license plate’s specific location is obtained.Test result proves that this algorithm locate rapidly,high accuracy and practical.

Key words: license plate location, weaken crest and fill trough, adaptation, edge detection

摘要: 提出一种针对汽车牌照定位的削峰填谷自适应算法。该算法首先对获取的彩色图像进行灰度化、灰度拉伸、边缘检测等处理,以突出车牌所在区域;然后对检测结果进行垂直投影,针对投影进行削峰填谷的自适应操作,获得车牌所在区域的左右边界;再对左右边界内的图像进行水平投影,并对投影进行削峰填谷的自适应操作,获得车牌所在区域的上下边界,从而最终确定车牌在原图像中的具体位置。实验结果表明该方法定位速度快、准确率高,实用性强。

关键词: 牌照定位, 削峰填谷, 自适应, 边缘检测

CLC Number: