Abstract
Lip-reading and lip identification are the hot spots in the technology of human-machine interaction. Researchers have been committed to work out the key issue that how to extract the most effective features for lip reading. This paper selects feature points and feature parameters for lip-reading based on FDP and FAP defined by MPEG-4. This paper uses an improved snake model on the location and extraction of the feature points. The algorithm has better robustness and has improved the accuracy of contour extraction. Then the SFFS method and correlation analysis are exploited to obtain the quicker optimal sub-aggregate and eliminate the redundancy features. Experimental results of the real data show that the features extracted from the sequence image have a high recognition rate in lip-readings recognition system.
This work is supported by pre-study fund of PLA Univ. Of Sci. & Tech(No.20110504).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Chiou, G.I., Hwang, J.N.: Lipreading by using snakes,principal component analysis and hidden Markov models to recognize color motion video. IEEE Trans. on Image Processing 6(8), 1192–1195 (1997)
Yao, H.-X., Liu, M.-B., Gao, W.: Method of Face Locating and Tracking Based on Chromatic Coordinates Transformation of color Images. Chinese J. Computers 23(2), 158–165 (2000) (in Chinese)
Potamianos, G., Graf, H.P., Cosatto, E.: An Image Transform Approach for HMM Based Aotumatic Lipreading. In: Proceeding of the International Conference on Image Processing, Chicagao, vol. III, pp. 173–177 (1998)
ISO/IEC 14496-1 Systems (2001)
ISO/IEC 14496-2 Visual (2001)
ISO/IEC 14496-2 Audio (2001)
Yang, Y., Zhang, T.-W.: A New Active Contour Tracking Algorithm. Chinese J. Computers 21, 297–302 (1998) (in Chinese)
Sun, Y., Li, J.: Landmine Detection Using Forward-Looking Ground Penetrating Radar. In: Proc. of SPIE, vol. 5794, pp. 1089–1097 (2005)
Sun, X., An, D., Liu, X.: Arbitray Lip Deformation Driven by Bopomofo Text. Journal of Computer-Aided Desin & Computer Graphics 20(12), 1063–1068 (2008) (in Chinese)
Sun, Y., Li, J.: Adaptive Learning Approach to Landmine Detection. IEEE Transactions on Aerospace and Electronic Systems 41(3), 973–985 (2005)
Yang, Y.-G., Zhou, Z.-M., Song, Q.: A nove feature evaluation and selection method for sequence images of VFGPIR. Signal Processing 25(10), 1574–1581 (2009) (in Chinese)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Rong, C., Yue, Z. (2011). A Novel Feature Selection and Extraction Method for Sequence Images of Lip-Reading. In: Lee, G. (eds) Advances in Automation and Robotics, Vol. 2. Lecture Notes in Electrical Engineering, vol 123. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-25646-2_46
Download citation
DOI: https://doi.org/10.1007/978-3-642-25646-2_46
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-25645-5
Online ISBN: 978-3-642-25646-2
eBook Packages: EngineeringEngineering (R0)