摘要
文本是计算机视觉的许多应用中的一项重要特征。大量复杂图像文本的应用 ,使图像文本分析技术成为研究的新方向。图像文本分析技术和通常的文档图像分析技术之间有着紧密的联系 ;但是图像文本其自身所具有的特性 ,又使得图像文本分析技术具有不同于一般的文档图像分析的更加丰富的内容。我们将图像文本分析技术划分为三大组成部分 :图像文本定位、图像文本的预处理和图像文本的识别进行讨论。最后 ,本文也对图像文本分析技术的应用进行了讨论。
Text is an important feature in applications of computer vision,especially for those related to information retrieval.Text embedded in complicated images,especially color images,is more and more often encountered.This brings birth to techniques focusing on image text analysis and their applications.Image text analysis cannot be viewed as an entirely independent technique to traditional document image analysis techniques.But,image text itself has its own distinct properties and rich contents compared with general document images.In this paper,techniques for image text analysis are divided into three main blocks:image text extraction,image text preprocessing and image text recognition.Promising application directions for image text analysis techniques are also presented.
出处
《计算机应用与软件》
CSCD
北大核心
2005年第2期76-78,94,共4页
Computer Applications and Software
关键词
计算机视觉
图像文本分析
图像处理
图像识别
Image text analysis Image text extraction Image text preprocessing Image text recognition