计算机与现代化 ›› 2025, Vol. 0 ›› Issue (06): 92-100.doi: 10.3969/j.issn.1006-2475.2025.06.015

• 图像处理 • 上一篇    下一篇

基于字符相似度的车牌识别多结果投票融合方法

  

  1. (1.中山大学智能工程学院,广东 深圳 518107; 2.广东省智能交通系统重点实验室,广东 深圳 518107;
    3.广州航海学院信息与通信工程学院,广东 广州 510725; 4.广东省公安厅,广东 广州 510050) 
  • 出版日期:2025-06-30 发布日期:2025-07-01
  • 作者简介: 作者简介:陈泽(1999—),男,广东湛江人,硕士研究生,研究方向:视频图像大数据分析,E-mail: chenz9@mail2.sysu.edu.cn; 通信作者:李熙莹(1972—),教授,博士,研究方向:基于视频图像的车辆检测与识别,交通视频大数据处理,E-mail: stslxy@mail.sysu.edu.cn; 通信作者:江倩殷(1987—),女,博士,研究方向:交通图像处理,机器学习,E-mail: qyjiang19@163.com; 林群雄(1974—),男,博士,研究方向:视频图像智能化技术,E-mail: linqx_zngcxy@163.com; 孙全忠(1975—),男,本科,研究方向:检测技术与自动化,图像处理,E-mail: kjksqz@163.com。
  • 基金资助:
    基金项目:公安部科技计划资助项目(2021JC38);广东省自然科学基金资助项目(2022A1515010361)

Multi-Result Voting Fusion Method for License Plate Recognition Based on Character Similarity

  1. (1. School of Intelligent Systems Engineering, Sun Yat-sen University, Shenzhen 518107, China; 
    2. Guangdong Province Key Laboratory of Intelligent Transportation System, Shenzhen 518107, China;
    3. School of Information and Communication Engineering, Guangzhou Maritime University, Guangzhou 510725, China;
    4. Guangdong Provincial Public Security Department, Guangzhou 510050, China)
  • Online:2025-06-30 Published:2025-07-01

摘要:
摘要:车牌识别是车辆信息识别的重要环节,但面临着相似字符识别混淆、识别结果的可信程度难以衡量等问题。为了提高车牌识别算法的可用性,本文提出一种基于字符相似度的车牌识别多结果投票融合方法。该方法根据字符相似度与贝叶斯定理推导车牌识别结果的投票权重,并选择投票权重最大的车牌识别结果作为投票结果。在计算投票权重的过程中,根据车牌字符的特点,提出以车牌整体为投票单元的整牌加权投票法与以单个车牌字符为投票单元的字符加权投票法。同时,本文计算投票结果的理论最小投票准确率,并作为投票结果置信度,用以衡量投票结果的可信程度。实验结果表明本文方法具有较好的效果,其投票准确率与易错字符投票准确率相比于传统加权投票法分别提高了0.78个百分点与2.18个百分点,能有效减少字符识别混淆的情况。同时表明了本文方法的有效性与稳定性,相比于传统加权投票方法,能更好地反映出投票结果的可信程度。



关键词: 关键词:字符相似度, 加权投票法, 结果融合, 车牌识别

Abstract:


Abstract: License plate recognition (LPR) is an important tool for vehicle information recognition. However, it faces some problems such as similar character confusion and unmeasurable credibility of recognition results. In order to improve the usability of LPR algorithms, this paper proposes a multi-result voting fusion method based on character similarity. The method calculates the voting weight of the LPR results according to the character similarity and Bayes theorem, and selects the largest voting weight as the voting result. In the process of calculating the voting weight, the plate weighted voting method and character weighted voting method are proposed according to the characteristics of license plate. Between them, the plate weighted voting method takes the whole license plate as the voting unit, and the character weighted voting method takes a single license plate character as the voting unit. At the same time, voting result confidence is calculated to measure the credibility of the voting results, and it is the theoretical minimum voting accuracy. The experimental results show that the proposed method can effectively reduce the confusion of character recognition. Compared with the traditional weighted voting method, the voting accuracy and error-prone character voting accuracy of the proposed method are increased by 0.78 percentage points and 2.18 percentage points respectively. The experimental results also show that the proposed voting result confidence calculation method is effective and stable. Compared with the traditional weighted voting method, the proposed method can better reflect the credibility of the voting results. 

Key words: Key words: character similarity, weighted voting, result fusion, license plate recognition

中图分类号: