Font adaptation of an HMM-based OCR system

被引：0
|
作者：
Ait-Mohand, Kamel [1 ]
Heutte, Laurent [1 ]
Paquet, Thierry [1 ]
Ragot, Nicolas [2 ]
机构：
[1] Univ Rouen, LITIS EA 4108, Ave Univ,BP 8, F-76801 St Etienne, France
[2] Univ Francois Rabelais Tours, F-37200 Tours, France
来源：
DOCUMENT RECOGNITION AND RETRIEVAL XVII | 2010年 / 7534卷
关键词：
OCR; font adaptation; MAP; MLLR; HMM;
D O I：
10.1117/12.840321
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
We create a polyfont OCR recognizer using HMM (Hidden Markov models) models of character trained on a dataset of various fonts. We compare this system to monofont recognizers showing its decrease of performance when it is used to recognize unseen fonts. In order to fill this gap of performance, we adapt the parameters of the models of the polyfont recognizer to a new dataset of unseen fonts using four different adaptation algorithms. The results of our experiments show that the adapted system is far more accurate than the initial system although it does not reach the accuracy of a monofont recognizer.
引用
收藏
页数：8
相关论文
共 50 条
  • [1] An experimental HMM-based postal OCR system
    Kornai, A
    [J]. 1997 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS I - V: VOL I: PLENARY, EXPERT SUMMARIES, SPECIAL, AUDIO, UNDERWATER ACOUSTICS, VLSI; VOL II: SPEECH PROCESSING; VOL III: SPEECH PROCESSING, DIGITAL SIGNAL PROCESSING; VOL IV: MULTIDIMENSIONAL SIGNAL PROCESSING, NEURAL NETWORKS - VOL V: STATISTICAL SIGNAL AND ARRAY PROCESSING, APPLICATIONS, 1997, : 3177 - 3180
  • [2] An HMM-based OCR for Persian/Arabic texts
    Ahmadi, A
    Omatu, S
    Yoshioka, M
    [J]. KNOWLEDGE-BASED INTELLIGENT INFORMATION ENGINEERING SYSTEMS & ALLIED TECHNOLOGIES, PTS 1 AND 2, 2001, 69 : 824 - 828
  • [3] Arabic OCR system analogous to HMM-based ASR systems; Implementation and evaluation
    Rashwan, M.A.
    Fakhr, M.W.
    Attia, M.
    EL-Mahallawy, M.S.
    [J]. Journal of Engineering and Applied Science, 2007, 54 (06): : 653 - 672
  • [4] Unsupervised adaptation for HMM-based speech synthesis
    King, Simon
    Tokuda, Keiichi
    Zen, Heiga
    Yamagishi, Junichi
    [J]. INTERSPEECH 2008: 9TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2008, VOLS 1-5, 2008, : 1869 - +
  • [5] Script-independent, HMM-based text line finding for OCR
    Lu, ZD
    Schwartz, R
    Raphael, C
    [J]. 15TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 4, PROCEEDINGS: APPLICATIONS, ROBOTICS SYSTEMS AND ARCHITECTURES, 2000, : 551 - 554
  • [6] Learning new acoustic events in an HMM-based system using MAP adaptation
    Geiger, Juergen T.
    Lakhal, Mohamed Anouar
    Schuller, Bjoern
    Rigoll, Gerhard
    [J]. 12TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2011 (INTERSPEECH 2011), VOLS 1-5, 2011, : 300 - 303
  • [7] A hierarchical, HMM-based automatic evaluation of OCR accuracy for a digital library of books
    Feng, Shaolei
    Manmatha, R.
    [J]. OPENING INFORMATION HORIZONS, 2006, : 109 - +
  • [8] FACTORED MLLR ADAPTATION FOR HMM-BASED EXPRESSIVE SPEECH SYNTHESIS
    Sung, June Sig
    Hong, Doo Hwa
    Lee, Chul Min
    Kim, Nam Soo
    [J]. 13TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2012 (INTERSPEECH 2012), VOLS 1-3, 2012, : 974 - 977
  • [9] An acoustic model adaptation using hmm-based speech synthesis
    Tanaka, K
    Kuroiwa, S
    Tsuge, S
    Ren, F
    [J]. 2003 INTERNATIONAL CONFERENCE ON NATURAL LANGUAGE PROCESSING AND KNOWLEDGE ENGINEERING, PROCEEDINGS, 2003, : 368 - 373
  • [10] Hmm-based system for transcribing Chinese handwriting
    Su, Tong-Hua
    Zhang, Tian-Wen
    Qiu, Zhao-Wen
    [J]. PROCEEDINGS OF 2007 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-7, 2007, : 3412 - +
12345