基于机器视觉的路牌文字识别方法研究  

Research on the Method of Street Sign Text Recognition Based on Machine Vision

作　　者：蒋润雨 张立珍[1] 徐聪 纪和俊 Jiang Runyu;Zhang Lizhen;Xu Cong;Ji Hejun(School of Electrical and Computer Engineering,Southeast University Chengxian College,Nanjing,China)

机构地区：[1]东南大学成贤学院电子与计算机工程学院,江苏南京

出　　处：《科学技术创新》2025年第9期97-100,共4页Scientific and Technological Innovation

摘　　要：随着人工智能的发展,越来越多的行业被人工智能所替代,我们经常会在走廊上看到一个又一个的送餐机器人,机器人在送餐过程中会按照指定路线行走。于是我们想通过机器视觉,通过捕获的图像对门牌或路牌进行文字识别。所以基于机器视觉的路牌文字识别应运而生。本文基于市场对路牌文字识别的需求,使用树莓派为核心处理器。在此基础上构建了一个小型的基于机器视觉的路牌文字识别系统。这个系统包括一个树莓派模块,图像获取模块以及图像处理和识别模块。电源选择为5 V输出电源,摄像头模块使用索尼FS02V1工业摄像头,识别模块使用了Tesseract-OCR,通过实验研究测试,本系统有可以灵活识别不同字体,精确度高,应用场景较广泛等特点。We see food delivery robots after another in the corridor,the robot will walk in accordance with the designated route during the delivery,So we want to use machine vision,through the captured images of the door or street sign for text recognition.So sign text recognition was born.We use Raspberry PI as the core processor.On this basis,a small machine vision based road sign character recognition system is constructed.The system includes a Raspberry PI,an image acquisition an image processing and recognition module.Use 5V output power supply,chose SONY FS02V1 camera,and the recognition module uses Tesseract-OCR.Through experimental research and testing,This system can recognize text flexibly and accurately.

关 键 词：机器视觉 文字识别 树莓派 Tesseract 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]