This paper presents an automatic lip reading system to increase security. The system depends mainly on recognizing the speaker using the visual speech signal alone. The proposed scheme works in two stages: the setting and the verification stage. At the setting stage, the system requests the user to utter a selected password. A video recording of the user face is captured and processed by a special words-based VSR system, which extracts a sequence of feature vectors. The features of each word are represented by a matrix. Each row of which represents a frame of the video, associated with the given word. Those features represent some lip region measurements, which include width, height, mutual information between successive frames, image quality measures, horizontal/vertical features ratio, teeth and tongue appearance. The sequence of the extracted features is stored on the authentication server.
In the verification stage, the same procedure is executed to extract the sequence of feature vectors for the word(s) uttered by the claimant. The system encrypts and sends the features to the authentication server. On receipt, the server decrypts and compares with the stored visual password feature vectors. The proposed scheme has been evaluated using a video database of 20 different speakers (10 females and 10 males), and different experiment sets. The evaluation has shown promising results, where the average equal error rate varies between 0.0% and 54.75% depending on the speaker. For this to become a practical approach, we need to do more investigation and experiments.
1st Mosharaka International Conference on Biomedical Engineering, Electronics and Nanotechnology (MIC-BEN 2011)
Congress
2011 Global Congress on Biomedical Engineering, Electronics and Nanotechnology (GC-BEN 2011), 9-11 September 2011, Amman, Jordan
Pages
11-16
Topics
Pattern Recognition Biometrics
ISSN
2227-331X
DOI
BibTeX
@inproceedings{177BEN2011,
title={Automatic lip reading for security},
author={Ahmad Hassanat},
booktitle={2011 Global Congress on Biomedical Engineering, Electronics and Nanotechnology (GC-BEN 2011)},
year={2011},
pages={11-16},
doi={}},
organization={Mosharaka for Research and Studies}
}