«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《集美大学学报（自然科学版）》[ISSN:1007-7405/CN:35-1186/N]

卷:

第18卷

期数:

2013年第5期

页码:

382-386

栏目:

数理科学与信息工程

出版日期:

2013-09-25

文章信息/Info

Title:

HSV-Based Selection Algorithm for Candidate License Plate Regions

作者:

杨奇¹; 吴云东²; 陈水利²

(1．集美大学理学院,福建 厦门361021;2．集美大学影像信息工程技术研究中心,福建 厦门361021)

Author(s):

YANG Qi¹; WU Yun-dong²; CHEN Shui-li²

(1 School of Science, Jimei University, Xiamen 361021,China;2.Research Centre of Image Information Engineering,Jimei University ,Xiamen 361021,China)

关键词:

车牌; 定位; HSV颜色空间; 平均密度

Keywords:

license plate; location; HSV color space; mean density

分类号:

-

DOI:

-

文献标志码:

-

摘要:

    提出了在HSV颜色空间中，利用车牌区域的颜色特征，筛选车牌区域的新算法.该算法根据车牌4种不同底色对应的3个分量，分别统计出底色平均密度的阈值范围，并利用该范围，对候选区域进行筛选，以确定车牌区域.在筛选车牌区域时，重点讨论了车身颜色与车牌底色相近或相同、候选车牌区域的纹理和边缘特征及宽高比都接近的情况.实验结果表明，该算法能较好地对车牌进行准确定位

Abstract:

It was proposed to selection algorithm for candidate license plate regions using color feature of vehicle license plate in HSV color space．Firstly，mean density threshold range of the background of vehicle license plate was calculated by using four kinds of background color of vehicle license plate corresponding to the three component threshold range.Secondly，according to the threshold range，the true vehicle license plate area was picked．Finally,it was discussed mainly for similar or identical of the color of car body and licese plate background ，closer of the texture、edge feature and aspect ratio of candidate licese plate regions when selecting license plate regions ．The experimental results showed that the proposed algorithm could efficiently detect vehicle license plate accurately．

参考文献/References:

-

相似文献/References:

[1]杨素珍.基于激光传感器定位的非完整AGV导引控制[J].集美大学学报(自然科学版),2018,23(5):373.
　YANG Suzhen.Guidance Control of Nonholonomic AGV with Location Based on Laser Sensor[J].Journal of Jimei University,2018,23(5):373.

备注/Memo

备注/Memo:

-

常用功能

更新日期/Last Update: 2014-06-28