Abstract
In this paper we introduce a mixture-state document segmentation method based on wavelet and the hidden Markov tree (HMT) models. First we propose a three-state HMT segmentation method that is similar to those in the reference [1]. Then through comparing the difference weights to the three-density Gaussian mixture distribution of different textures, we find that background, text and image can be well approximated respectively by one-state and two-state and three-state HMT models. Then we get a new segmentation method, mixture-state HMT segmentation. Experiments with scanned document images indicate that the new approach improves the segmentation accuracy over the raw segmentation in [1].
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
H. Choi and R. G. Baraniuk, Multiscale Document Segmentation using Wavelet-Domain Hidden Markov Models, Science & Technology, Janu. 2000.
M. S. Crouse, R. D. Nowak, and R. G. Baraniuk, “Wavelet-Based Statistical Signal Processing using Hidden Markov Models,” IEEE Trans. Signal Proc. 46, April 1998.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Tang, Y.Y., Hou, Y., Song, J., Yang, X. (2001). Mixture-State Document Segmentation Using Wavelet-Domain Hidden Markov Tree Models. In: Tang, Y.Y., Yuen, P.C., Li, Ch., Wickerhauser, V. (eds) Wavelet Analysis and Its Applications. WAA 2001. Lecture Notes in Computer Science, vol 2251. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45333-4_29
Download citation
DOI: https://doi.org/10.1007/3-540-45333-4_29
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-43034-6
Online ISBN: 978-3-540-45333-8
eBook Packages: Springer Book Archive