印刷体英文OCR系统的研究与实现  

Research and Implementation of Printed English OCR System

作　　者：蒋东玉 田英鑫[2] 

机构地区：[1]哈尔滨华德学院电子与信息工程学院,哈尔滨150025 [2]哈尔滨工业大学软件学院,哈尔滨150001

出　　处：《智能计算机与应用》2014年第4期111-112,117,共3页Intelligent Computer and Applications

摘　　要：OCR是一种文字自动识别技术,能够将图像中的文字输入到计算机中,属于非键盘输入范畴,使用到的图像输入设备主要是扫描仪。本文探讨了印刷体英文OCR的实现过程。通过对图像预处理,字符分割,字符识别三个主要模块的分析,使用了便于初学者理解的算法来实现各模块的功能。设计清晰,实现简单,而且具有较快的识别速度和可接受的准确率。OCR is an automatic text recognition technology, which could input the text in the image into the computer. OCR can be attributed to non - keyboard input areas, using the image input device is mainly scanner. This paper discusses the implementation process of the OCR printed in English. Through the image preprocessing, character segmentation, character recognition analysis of three main modules, the algorithm easy for beginners to understand is used to achieve the function of each module. The design is clear, simple, and has a faster recognition speed and acceptable accuracy.

关 键 词：OCR技术 字符分割 字符识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]