基于局部投影方差的快速车牌校正方法
DOI:
作者:
作者单位:

浙江工业大学计算机科学与技术学院,,,,浙江工业大学计算机科学与技术学院

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金资助项目(C12412135, 61402410); 浙江省重点研发计划项目(2018C01064)


A fast correction method for vehicle license plate based on local projection variance
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    车牌倾斜与错切校正作为车牌识别的重要环节,虽已提出不少方法,但仍存在准确度低、效率低等不足。鉴于现有方法的不足提出一种高效的车牌倾斜与错切校正方法。首先,根据车牌图像的投影方差存在着明显的局部性规律,提取车牌的局部特征图像用于角度检测。并提出一种基于坐标旋转的最大投影方差获取方法,快速得到车牌水平倾斜角度和垂直错切角度。此外,针对字符本身的结构对于垂直错切角度检测的影响,提出相应的解决策略。最终使用3000张情况各异的车牌进行实验。实验结果表明,在正确率比现有传统方法略高的情况下,算法效率也得到大幅提升。

    Abstract:

    As an important part of license plate recognition, a lot of tilt and shear correction methods of license plate have been put forward. However, there are still some deficiencies. In order to improve the efficiency and accuracy of the tilt and shear correction methods of vehicle license plate, a series of improvement measures are put forward. First of all, special parts are extracted from the vehicle license plate to be used for horizontal tilt correction and vertical shear correction respectively. Then, according to the distribution rule of the projection variance, the angle with the maximum projection variance in the horizontal direction and vertical direction will be found respectively. These two angles are corresponding to the horizontal tilt angle and vertical shear angle of the vehicle license plate. In order to eliminate the influence of the structure of characters, the corresponding strategy is proposed. Experiments were conducted on 3000 real vehicle license plates in different situations. The experimental results show that not only the accuracy of the proposed method is slightly higher than the existing traditional methods, but also the efficiency has been greatly improved.

    参考文献
    相似文献
    引证文献
引用本文

汪敏倩,卢书芳,张元鸣,张永良,高飞.基于局部投影方差的快速车牌校正方法计算机测量与控制[J].,2019,27(1):228-232.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2018-07-05
  • 最后修改日期:2018-07-25
  • 录用日期:2018-07-25
  • 在线发布日期: 2019-01-25
  • 出版日期: