一种免除二值化的视频叠加中文字符识别方法

doi:10.7523/j.issn.2095-6134.2018.03.015

中国科学院大学学报 ›› 2018, Vol. 35 ›› Issue (3): 402-408.DOI: 10.7523/j.issn.2095-6134.2018.03.015

一种免除二值化的视频叠加中文字符识别方法

田洁, 王伟强, 孙翼

中国科学院大学计算机与控制学院, 北京 101408

收稿日期:2017-03-15 修回日期:2017-04-19 发布日期:2018-05-15
通讯作者: 田洁
基金资助:
国家自然科学基金（61271434）资助

Recognition of overlaid Chinese characters in videos without binarization

TIAN Jie, WANG Weiqiang, SUN Yi

School of Computer and Control Engineering, University of Chinese Academy of Sciences, Beijing 101408, China

Received:2017-03-15 Revised:2017-04-19 Published:2018-05-15

摘要/Abstract

摘要： 提出一种新的用于识别视频中字幕文字的方法。鉴于视频中文字的大小、颜色、渲染风格和分辨率的不同，以及视频中各种复杂背景的影响，识别视频中的叠加文字是一个尚未解决的问题。目前，大多数视频叠加文字识别方法都基于视频文字的二值化和传统OCR引擎的结合。然而，二值化过程容易引入噪声和文字笔划信息的丢失。另外，传统OCR技术主要专注于高分辨率的扫描打印文档，这些文档具有背景单一、噪声少和笔划信息较完整的特点。因此，传统OCR引擎用于识别叠加文字二值化后的结果可能不够鲁棒。为解决这个问题，直接从未二值化的叠加视频文字图像中提取Gabor特征用于训练二层字符识别器。实验结果表明，本文提出的方法在多字体视频叠加中文文字识别上有良好的效果。

关键词: 视频叠加文字, OCR, Gabor, 最近原型分类(NPC)

Abstract: In this paper, a new method for recognizing caption texts in videos is proposed. Due to varying font sizes, colors, styles, and resolutions and complex backgrounds in videos, it is still a challenging problem to recognize overlaid texts in videos. Most existing overlaid text recognition methods are based on the combination of text binarization and traditional OCR engine. However, the process of text binarization may incur noises and text stroke information loss. Additionally, techniques of traditional OCRs are mainly focused on high-resolution scans of printed documents, which have the characteristics of single color background, little noise, and more complete stroke information. Hence, traditional OCR engines might not be robust enough to recognize the binarization results of overlaid text images. In order to solve this problem, we directly extract Gabor features from overlaid text images without binarization for training the two-level character recognizer. The final experimental results demonstrate that the proposed method makes a great progress in overlaid Chinese text recognition with multiple fonts.

Key words: video overlaid text, OCR, Gabor, nearest prototype classifier

中图分类号:

TP393

田洁, 王伟强, 孙翼. 一种免除二值化的视频叠加中文字符识别方法[J]. 中国科学院大学学报, 2018, 35(3): 402-408.

TIAN Jie, WANG Weiqiang, SUN Yi. Recognition of overlaid Chinese characters in videos without binarization[J]. , 2018, 35(3): 402-408.

参考文献

[1] Yang H, Meinel C. Content based lecture video retrieval using speech and video text information[J]. IEEE Transactions on Learning Technologies, 2014, 7(2):142-154.
[2] Jung K, Kim K I, Jain A K. Text information extraction in images and video:a survey[J]. Pattern Recognition, 2004, 37(5):977-997.
[3] Wang Z, Yang L, Wu X, et al. A survey on video caption extraction technology[C]//Multimedia Information Networking and Security (MINES), 2012 Fourth International Conference on. IEEE, 2012:713-716.
[4] Yusufu T, Wang Y, Fang X. A video text detection and tracking system[C]//Multimedia (ISM), 2013 IEEE International Symposium on. IEEE, 2013:522-529.
[5] Lienhart R W, Stuber F. Automatic text recognition in digital videos[C]//Electronic Imaging:Science & Technology. International Society for Optics and Photonics, 1996:180-188.
[6] Niblack W. An introduction to digital image processing[M]. Birkeroed:Strandberg Publishing Company, 1985.
[7] Sauvola J, Pietikäinen M. Adaptive document image binarization[J]. Pattern Recognition, 2000, 33(2):225-236.
[8] Wolf C, Jolion J M, Chassaing F. Text localization, enhancement and binarization in multimedia documents[C]//Pattern Recognition, 2002. Proceedings. 16th International Conference on. IEEE, 2002, 2:1037-1040.
[9] Huang X, Ma H, Zhang H. A new video text extraction approach[C]//Multimedia and Expo, 2009. ICME 2009. IEEE International Conference on. IEEE, 2009:650-653.
[10] Kita K, Wakahara T. Binarization of color characters in scene images using k-means clustering and support vector machines[C]//Pattern Recognition (ICPR), 201020th International Conference on. IEEE, 2010:3183-3186.
[11] Peng X, Setlur S, Govindaraju V, et al. Markov random field based binarization for hand-held devices captured document images[C]//Proceedings of the Seventh Indian Conference on Computer Vision, Graphics and Image Processing. ACM, 2010:71-76.
[12] Wang Y, Shi C, Xiao B, et al. MRF based text binarization in complex images using stroke feature[C]//Document Analysis and Recognition (ICDAR), 201513th International Conference on. IEEE, 2015:821-825.
[13] Mancas-Thillou C, Gosselin B. Character segmentation-by-recognition using log-Gabor filters[C]//Pattern Recognition, 2006. ICPR 2006.18th International Conference on. IEEE, 2006, 2:901-904.
[14] Huang X. A novel video text extraction approach based on Log-Gabor filters[C]//Image and Signal Processing (CISP), 20114th International Congress on. IEEE, 2011, 1:474-478.
[15] Mishra A, Alahari K, Jawahar C V. Unsupervised refinement of color and stroke features for text binarization[J]. International Journal on Document Analysis and Recognition, 2017,20(2):1-17.
[16] Hao Q, Feng Z D, Ge Y. A study on the use of Gabor features for Chinese OCR[C]//Intelligent Multimedia, Video and Speech Processing, 2001. Proceedings of 2001 International Symposium on. IEEE, 2001:389-392.
[17] Wang X, Ding X, Liu C. Gabor filters-based feature extraction for character recognition[J]. Pattern recognition, 2005, 38(3):369-379.
[18] Jin X B, Liu C L, Hou X. Regularized margin-based conditional log-likelihood loss for prototype learning[J]. Pattern Recognition, 2010, 43(7):2428-2438.
[19] Smith R. An overview of the Tesseract OCR engine[C]//Document Analysis and Recognition, 2007. ICDAR 2007. Ninth International Conference on. IEEE, 2007, 2:629-633.
[20] Otsu N. A threshold selection method from gray-level histograms[J]. Automatica, 1975, 11(285-296):23-27.
[21] Zhang Z, Wang W, Lu K. Video text extraction using the fusion of color gradient and log-Gabor filter[C]//Pattern Recognition (ICPR), 201422nd International Conference on. IEEE, 2014:2938-2943.

一种免除二值化的视频叠加中文字符识别方法

Recognition of overlaid Chinese characters in videos without binarization

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价

访问统计

联系我们