- International Journal of Information Security Science
- Vol: 9 Issue: 2
- Multi-Modal Biometrics Fusion Based on Component Analysis and Stationery Wavelet Transform
Multi-Modal Biometrics Fusion Based on Component Analysis and Stationery Wavelet Transform
Authors : Gabriel Babatunde Iwasokun, Kolawole Isaiah Opatoye, Babajide Oluwaseun Orunmuyi
Pages : 114-125
View : 14 | Download : 5
Publication Date : 2020-06-01
Article Type : Research
Abstract :It has been observed that the accuracy of multimodal biometric system is highly dependent on the adequacy of the applied fusion technique. Fusion at sample, template, matching and ranking levels have all proved reasonable contributions to the performance of the multi-modal systems. In this paper, a model that is based on the combination of Principal Component Analysis PCA and Stationary Wavelet Transform SWT is proposed for the fusion of biometric images. The model comprises image depuration, histogram balancing, pruning and homogenization as well as PCA-based feature extraction stages. The decomposition and fusion of the images using the extracted features were based on SWT. The experimental study of the model with standard face and ear images revealed its suitability for obtaining high quality fusion. The obtained Peak Signal to Noise Ratio PSNR , Mean Square Error MSE and Standard Deviation SD values established the superiority of the proposed model over some related onesKeywords : Multimodal biometrics, face, ear, biometric fusion, filtering, PCA, SWT