Development of an intelligent multimodal biometric system for household appliances control
Over the years, biometrics features which include but not limited to face, fingerprints, iris, hand geometry, palmprint have been explored as a means of individual/ personal recognition. However, the existing spectra of biometric devices in real world application are contact-based devices and are mo...
| Main Author: | |
|---|---|
| Format: | Monograph |
| Language: | English English |
| Published: |
[s.n]
2012
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/31216/ http://irep.iium.edu.my/31216/1/End_of_Project_Report_Form_EDW_B11-012-490.pdf http://irep.iium.edu.my/31216/2/Full_Final_Report_EDW_B11-012-490.pdf |
| Summary: | Over the years, biometrics features which include but not limited to face, fingerprints, iris, hand geometry, palmprint have been explored as a means of individual/ personal recognition. However, the existing spectra of biometric devices in real world application are contact-based devices and are mostly unimodal systems. These systems have not only generated issues on public health but also are susceptible to several problems such as high intra-class and interclass variations. It may be possible to enhance the multi-functionalism of these systems by integrating two or more biometrics. Hence, in this research, multimodal biometric system involving the integration of palmprint, teeth and voice biometrics was developed. The development of the multimodal biometric system was conducted in two phases. In the first phase, individual biometrics recognition algorithms were developed for each biometrics feature. Palmprint recognition system in unrestricted environment requires a robust algorithm capable of adapting to complex environment. To this effect, a novel skin segmentation algorithm using artificial neural network (ANN) technique was developed and this was deployed to segment the palmprint and face images from the background. Similarly, a novel region of interest (ROI) extraction technique was developed from which the palmprint creases specifically the principal lines were extracted and teeth were then extracted from the segmented face images respectively. The extracted lines were characterized by discrete Fourier transform (DFT) coefficients of the detected K-endpoint distance matrix which was subsequently used for recognition. On the other hand, teeth recognition was based on the magnitudes of the extracted DFT coefficients of the teeth radii signature. Furthermore, a text-dependent voice recognition scheme was developed by extracting the vocal tract features of individuals’ on the pronunciation of a specific word using linear predictive coding (LPC) technique. The extracted coefficients were trained and evaluated for individual recognition using ANN paradigm. Having successfully developed a unimodal system for each biometric deployed, the features were then integrated in the second phase at the score level to obtain the proposed multimodal system used for home appliances control. In the unimodal systems, average recognition rates of 100%, 87.9% and 65.7% were obtained for palmprint, teeth and voice recognition systems respectively. After the evaluation of each unimodal systems, these systems were integrated at the score level to achieve the proposed multimodal biometric system. Evaluation of this system yielded an average recognition rate of 99.1%. The developed multimodal system was later applied for restricting unauthorized users from the operation of home appliances which was basically controlled by hand gesture recognition. The control mechanism achieved 100% accuracy. Experimental results of this scheme have demonstrated the possibility deployment in other areas such as mobile devices, computer security, and forensic security amongst others.
|
|---|