基于STM32的车牌识别系统的设计与开发
Design and Development of License Plate Recognition System Based on STM32
摘要: 随着我国机动车保有量持续攀升,停车场、道路交通监测系统对嵌入式交通设备的需求不断增长。本文针对现有车牌识别系统在嵌入式、小型化、低功耗方面的不足,提出了一种基于STM32微控制器的嵌入式车牌识别系统。该系统硬件设计以STM32F103RBT6嵌入式处理器为核心,集成OV7670图像传感器与LCD显示模块,车牌识别软件则部署于STM32F103RBT6处理器上。所设计的车牌识别系统通过对OV7670图像传感器采集的车牌图像数据进行定位、倾斜矫正、字符分割及匹配等过程,实现车牌图像字符的提取与识别。实验结果表明,本文所设计的系统在光照变化明显、车牌存在倾斜以及背景复杂干扰场景下,均具有较高的识别准确率,且功耗显著优于传统方案,为停车场管理、交通监控等场景提供了高性价比的嵌入式解决方案。
Abstract: With the continuous growth of the number of motor vehicles in China, the demand for embedded traffic equipment in parking lots and road traffic monitoring systems is increasing. Aiming at the shortcomings of existing license plate recognition systems in terms of embedding, miniaturization and low power consumption, this paper proposes an embedded license plate recognition system based on STM32 microcontroller. The hardware design of the system mainly uses the STM32F103RBT6 embedded processor as the core, integrating the OV7670 image sensor and LCD display module. The license plate recognition algorithm designed for the system is deployed on the STM32F103RBT6 processor. By analyzing the image data collected by the OV7670 image sensor and through modules such as license plate positioning, tilt correction, character segmentation and matching, the algorithm achieves accurate recognition of license plate characters. Experimental results show that the system designed in this paper exhibits high recognition accuracy in scenarios with significant changes in illumination, tilted license plates and complex vehicle backgrounds. Moreover, its power consumption is significantly better than that of traditional schemes, effectively balancing the system’s performance and energy consumption. This research provides a cost-effective solution for parking lot management, traffic monitoring and other scenarios, and verifies the feasibility of applying the STM32 platform in intelligent transportation systems.
文章引用:孟佳聪, 廖雨宁, 陈欣雨, 黄潇威, 张涵洁, 邓伟, 周诗源. 基于STM32的车牌识别系统的设计与开发[J]. 图像与信号处理, 2025, 14(4): 429-442. https://doi.org/10.12677/jisp.2025.144040

参考文献

[1] 王晓光, 王群. 用于车牌字符识别的SVM算法[J]. 现代电子技术, 2004(8): 8-10.
[2] 魏武, 黄心汉, 张起森, 等. 基于模板匹配和神经网络的车牌字符识别方法[J]. 模式识别与人工智能, 2001, 14(1): 123-127.
[3] 谭本军. 基于云服务的仓库防盗系统的设计与研究[D]: [硕士学位论文]. 长沙: 湖南大学, 2019.
[4] 张义显. 基于STM32的新型家庭报警器的设计与实现[D]: [硕士学位论文]. 秦皇岛: 燕山大学, 2016.
[5] 张少鹏. 基于STM32的车牌识别手持移动终端的设计[D]: [硕士学位论文]. 兰州: 西北师范大学, 2015.
[6] 朱志巍, 陈华东, 张洁. 基于STM32的输油管巡检平台设计[J]. 机器人技术与应用, 2018(5): 34-36.
[7] 王四伟. 基于SOPC的指纹采集与处理系统的研究[D]: [硕士学位论文]. 武汉: 武汉理工大学, 2009.
[8] 杨鼎鼎, 陈世强, 刘静漪. 基于车牌背景和字符颜色特征的车牌定位算法[J]. 计算机应用与软件, 2018, 35(12): 216-221.
[9] 程聃, 陆华才, 高文根. 基于改进Canny算子边缘检测和数学形态学的车牌定位算法[J]. 黑龙江工业学院学报(综合版), 2019, 19(12): 68-72.
[10] 杨俊, 戚飞虎. 一种基于形状和纹理特征的车牌定位方法[J]. 计算机工程, 2006(2): 170-171, 202.
[11] 林彬. 基于深度学习的车牌定位与识别研究[D]: [硕士学位论文]. 厦门: 厦门理工学院, 2023.
[12] 张国权, 李战明, 李向伟, 等. HSV空间中彩色图像分割研究[J]. 计算机工程与应用, 2010, 46(26): 179-181.