首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于文字特征的文档碎片拼接复原研究
引用本文:耿文欣.基于文字特征的文档碎片拼接复原研究[J].焦作大学学报,2014(2):93-95.
作者姓名:耿文欣
作者单位:长沙理工大学数学与计算科学学院,湖南长沙41000
摘    要:分析了利用图像拼接技术以及基于文档碎片几何特征的文档碎片自动拼接方法的缺点,提出了基于文字特征的文档碎片拼接复原的新方法。该方法首先将文档碎片图像数字化处理后得到灰度矩阵。提取文档碎片中的文字特征,获得文档碎片左侧白色灰度距离以及文档碎片中心距离向量:然后利用中心匹配法对文档碎片进行聚类分析得到同行文档碎片,并建立最小距离模型对同行文档碎片进行拼接,得到若干行碎片;最后再次利用最小距离模型将行碎片进行拼接.从而实现文档碎片的拼接复原。试验表明新方法提出的拼接算法真实可靠。

关 键 词:文字特征  灰度矩阵  中心匹配法  最小距离模型

Research on the Restoration of Document Fragments Based on Text Characteristics
GENG Wenxin.Research on the Restoration of Document Fragments Based on Text Characteristics[J].Journal of Jiaozuo University,2014(2):93-95.
Authors:GENG Wenxin
Institution:GENG Wenxin ( Changsha University of Science and Technology, Changsha 41000, China)
Abstract:The article analyzed the shortcomings of the image stitching technology and automatic mosaic method based on document fragment geometric features, and put forward a new method of document fragments stitching restoration based on the text features. The method first turned the document fragments into digital image processing to obtain gray matrix, extracted the text feature in the document fragments obtained the white gray distance on the left to the document fragments and the center distance vector document of the fragments; then with the center matching method to make an analysis of clustering document fragments to get the similar document fragments, and establish the minimum distance model for peer document fragment splicing to gain some row pieces; finally, again with the least distance model putting pieces of mosaic in order to achieve document fragments restoring. The test shows that the stitching algorithm that the new method advanced is reliable.
Keywords:text feature  gray matrix  center matching method  minimum distance model
本文献已被 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号