Wavelet transformation on hidden Markov models

被引：0
|
作者：
Wang, MH [1 ]
Ma, H [1 ]
机构：
[1] Sichuan Univ, Math Coll, Chengdu 610064, Peoples R China
来源：
WAVELET ANALYSIS AND ITS APPLICATIONS (WAA), VOLS 1 AND 2 | 2003年
关键词：
D O I：
暂无
中图分类号：
O29 [应用数学];
学科分类号：
070104 ;
摘要：
In this paper, a new method to adapt the hidden Markov model (HMM) parameters is proposed when the original feature vector sequences extracted from handwritten characters are transformed by wavelet. By our method, we are able to be free from the requirement of retraining the whole recognition parameters when the feature vectors are changed and makes it sufficient to adapt the parameters to the given impulse response of wavelet filter. We evaluate the performance of the proposed method, and compare it to that of the retrained HMM parameters based on the handwritten character recognition experiments.
引用
收藏
页码：361 / 367
页数：7
相关论文
共 50 条
  • [1] Hidden Markov models for wavelet image separation and denoising
    Ichir, MM
    Mohammad-Djafari, A
    [J]. 2005 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1-5: SPEECH PROCESSING, 2005, : 225 - 228
  • [2] Improved hidden Markov models in the wavelet-domain
    Fan, GL
    Xia, XG
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2001, 49 (01) : 115 - 120
  • [3] Combining wavelet transform and hidden Markov models for ECG segmentation
    Andreäo R.V.
    Boudy J.
    [J]. EURASIP Journal on Advances in Signal Processing, 2007 (1)
  • [4] Hidden Markov models for wavelet-based signal processing
    Crouse, MS
    Baraniuk, RG
    Nowak, RD
    [J]. THIRTIETH ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS, VOLS 1 AND 2, 1997, : 1029 - 1035
  • [5] Wavelet-based denoising using hidden Markov models
    Borran, MJ
    Nowak, RD
    [J]. 2001 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS I-VI, PROCEEDINGS: VOL I: SPEECH PROCESSING 1; VOL II: SPEECH PROCESSING 2 IND TECHNOL TRACK DESIGN & IMPLEMENTATION OF SIGNAL PROCESSING SYSTEMS NEURALNETWORKS FOR SIGNAL PROCESSING; VOL III: IMAGE & MULTIDIMENSIONAL SIGNAL PROCESSING MULTIMEDIA SIGNAL PROCESSING, 2001, : 3925 - 3928
  • [6] Combining wavelet transform and hidden Markov models for ECG segmentation
    Andreao, Rodrigo Varejao
    Boudy, Jerome
    [J]. EURASIP JOURNAL ON ADVANCES IN SIGNAL PROCESSING, 2007,
  • [7] Multiresolution wavelet analysis and hidden Markov models in biomedical engineering
    Bovbel, EI
    Kheidorov, IE
    Chaikov, YA
    [J]. ANALYSIS OF BIOMEDICAL SIGNALS AND IMAGES, PROCEEDINGS, 2002, : 203 - 205
  • [8] Rainfall data simulation by hidden Markov model and discrete wavelet transformation
    A. W. Jayawardena
    P. C. Xu
    W. K. Li
    [J]. Stochastic Environmental Research and Risk Assessment, 2009, 23 : 863 - 877
  • [9] Rainfall data simulation by hidden Markov model and discrete wavelet transformation
    Jayawardena, A. W.
    Xu, P. C.
    Li, W. K.
    [J]. STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT, 2009, 23 (07) : 863 - 877
  • [10] Markov models - hidden Markov models
    Grewal, Jasleen K.
    Krzywinski, Martin
    Altman, Naomi
    [J]. NATURE METHODS, 2019, 16 (09) : 795 - 796
12345