期刊文献+

基于极坐标转换的中文印章文字识别 被引量:5

Chinese seal text recognition based on polar coordinate conversion
下载PDF
导出
摘要 为能够处理文档中印章元素的关键信息,促进办公智能化,提出一种基于极坐标转换的印章文字识别方法。根据印章元素通常成圆环状排列的特点,对中文印章图像进行极坐标展开,克服印章文字方向不统一的问题,利用CTPN+CRNN网络进行文字的检测与识别,对CTPN网络中的文本构造算法进行改进,实现对印章内容快速准确的识别。用该算法对自制的中文印章数据集进行实验,印章内容的文字识别召回率可以达到90.4%,表明该算法可以有效检测识别印章内容,对文档的分类与鉴别研究具有重要的意义。 To process the key information of seal elements in documents and promote the intelligence of office,a method of seal character recognition based on polar coordinate transformation was proposed.According to the characteristics of seal elements that were usually arranged in circles,the polar coordinates of Chinese seal images were developed to overcome the problem of inconsistent seal character direction.CTPN+CRNN network was used for character detection and recognition,and the text construction algorithm in CTPN network was improved to realize the fast and accurate recognition of seal content.The algorithm was used on the self-made Chinese seal data set,the recall rate of the seal content can reach 90.4%,indicating that this algorithm can effectively detect and identify the seal content,which is of great significance to the classification and identification research of documents.
作者 戴俊峰 杨天 熊闻心 DAI Jun-feng;YANG Tian;XIONG Wen-xin(State Grid Hubei Information and Telecommunication Limited Company,Wuhan 430077,China;School of Electronic Information,Wuhan University,Wuhan 430072,China)
出处 《计算机工程与设计》 北大核心 2021年第11期3174-3180,共7页 Computer Engineering and Design
基金 国网湖北省电力有限公司科技基金项目(52153318004G) 国家自然科学基金项目(61471272)。
关键词 印章识别 极坐标转换 文档鉴别 文字分割 文字检测 seal recognition polar coordinate conversion document identification text segmentation text detection
  • 相关文献

参考文献2

二级参考文献22

  • 1戴声奎,喻莉,朱光喜,刘文予.基于视频时空相关性的帧内预测模式抉择[J].通信学报,2005,26(11):43-48. 被引量:4
  • 2DONOHO D L, IAIN M J. Ideal spatial adaptation by wavelet shrink- age[J]. Biometrika, 1994, 81(3): 425-455.
  • 3DONOHO D L. De-noising by soft- thresholding[J]. IEEE Trans on Information Theory, 1995, 41(3): 613- 627.
  • 4PIZURICA A, PHILIPS W, LEMAHIEU I, et al. A joint interand intrascale statistical model for Bayesian wavelet based image denois- ing[J]. IEEE Tram on Image Process, 2002, 11(5): 545-557.
  • 5SELESNICK I W, LI K Y. Video denoising using 2D and 3D dual-tree complex wavelet transforms[A]. In Wavelet Applications in Signal and Image Processing X (Proc SPIE 5207) [C]. San Diego, 2003.4-8.
  • 6VARGHESE C~ ZHOU W. Video danoising using a spatiotemporal statistical model of wavelet coefficients[A]. IEEE International Con- ference on Acoustics, Speech and Signal Prncessing[C]. Las Vegas, 2008.1257-1260.
  • 7KOSTADIN D, ALESSANDRO F, KAREN E. Video danoising by sparse 3D transform-domain collaborative filtering[A]. European Sig- nal Processing Conference[C]. Tampere, Finland, 2007.145 -149.
  • 8Yl H, LIU C Q, SI-IEN Z W, et al. Robust video denoising using low rank matrix completion[A]. IEEE Conference on Computer Vision andPattern Recognition (CVPR)[C]. San Francisco, CA,2010. 1791-1798.
  • 9CAMILLE S, CHARLES A D. Adaptive regularization of the nl-means application to image and video denoising[J]. IEEE Transac- tions on Image Processing, 2014,23(8): 3506 -3520.
  • 10KUANG Y, ZHANG L. An adaptive rank-sparsity K-SVD algorithm for image sequence denoising[J]. Pattern Recognition Letters, 2014, 45(11): 46-54.

共引文献18

同被引文献37

引证文献5

二级引证文献4

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部