Lip feature extraction based on Pulse Coupled Neural Network

Author

Wang, Mengjun ; Wang, Xiangling ; Li, Gang

Author_Institution

Sch. of Inf. Eng., HeBei Univ. of Technol., Tianjin, China

Volume

2

fYear

2011

fDate

15-17 Oct. 2011

Firstpage

924

Lastpage

927

Abstract

Pulse Coupled Neural Network (PCNN) is used to extract lip features in the gray image sequences of visual speech, Time series, Entropy series, Logarithm series, and Standard deviation are considered as the feature vector. Experiments are carried out based on HMM with 4 states and 16 Gaussian mixture components in a small database for speaker-dependent case. Comparing with the traditional feature extracting method by Discrete Cosine Transform (DCT), Experiment results show that feature vector based on PCNN get the higher recognition rates than feature vector based on DCT. The maximum recognition rate improves 7.87% than DCT based lip feature.

Keywords

Gaussian processes; entropy; feature extraction; hidden Markov models; image recognition; image sequences; neural nets; time series; Gaussian mixture components; HMM; PCNN; entropy series; feature vector; gray image sequences; hidden Markov model; lip feature extraction; logarithm series; pulse coupled neural network; recognition rates; standard deviation; time series; visual speech; Discrete cosine transforms; Entropy; Feature extraction; Hidden Markov models; Joining processes; Neurons; Vectors; Hidden Markov Model; PCNN; entropy sequence; feature vector; logarithmic sequence; normalized DCT coefficients; standard variance sequence; time series;

fLanguage

English

Publisher

ieee

Conference_Titel

Image and Signal Processing (CISP), 2011 4th International Congress on

Conference_Location

Shanghai

Print_ISBN

978-1-4244-9304-3

Type

conf

DOI

10.1109/CISP.2011.6100368

Filename

6100368